In an exclusive interview, Patankar explains how Protectt.ai is aiming to reshape the mobile app security landscape. What struck me most was Protectt.ai’s intense focus on real-time, in-app threat protection and the level of sophistication they bring to fighting the hidden – and sometimes not-so-hidden – adversaries lurking behind every suspicious link, tampered APK, or cleverly disguised piece of malware.

Providing constant mobile app security

Mobile security is often portrayed as a tug-of-war between developers who patch vulnerabilities and hackers who exploit them. Protectt.ai flips this dynamic on its head. Their approach is not just to block known threats, but to anticipate and neutralize them as they happen.

“At Protectt.ai, our core strengths in the mobile app security landscape are centered around our advanced Runtime Application Self-Protection (RASP) technology, which sets us apart from competitors,” Patankar explained. The mention of RASP (Runtime Application Self-Protection) particularly piqued my interest. Rather than relying solely on perimeter defenses, RASP technology defends the app from within, making it harder for attackers to manipulate code or data.

As Patankar put it, “We leverage sophisticated AI and ML algorithms for enhanced threat detection, allowing us to adapt quickly to evolving attack vectors.” It’s a sentiment that encapsulates the company’s approach: treat security as a living, evolving system, not a static checklist of defenses.

Also read: Cybersecurity in Age of AI: Black Hat 2024’s top 3 LLM security risks

Traditional mobile antivirus solutions rely on signature-based detection, which are great for known threats but less effective against zero-day exploits or entirely new malware species. Protectt.ai, on the other hand, uses a cloud-based analysis system and next-gen runtime capabilities.

“Our RASP solution provides continuous, in-app security, detecting and responding to threats as they occur, ensuring immediate mitigation of vulnerabilities,” Patankar told me. This is critical because a threat can appear out of nowhere – a malicious snippet of code injected into a supposedly benign update, or a cleverly disguised phishing attempt exploiting a user’s trust in a familiar brand.

By maintaining a scalable cloud infrastructure, Protectt.ai can correlate threat intelligence across vast data sets. According to Patankar, “Our scalable cloud infrastructure enables efficient threat intelligence and provides actionable insights to strengthen our clients’ mobile app security posture.”

How Protectt.ai is bridging the mobile app security gap

One might think that downloading apps only from official app stores like Google Play or Apple’s App Store provides ironclad safety. While these platforms do perform initial security scans, Patankar was quick to point out that their checks aren’t foolproof.

“Google Play Store conducts an initial security check before publishing apps, but this vetting process can still leave gaps, especially with advanced threats that evolve over time,” he explained. The truth is, attackers don’t stand still; they evolve. They reverse-engineer apps, tamper with code, and distribute malicious versions via phishing links. Once installed, these rogue apps can compromise user data or even carry out unauthorized financial transactions.

“Unlike Google Play’s static vetting, AppProtectt continuously monitors apps in real-time, detecting and neutralizing dynamic threats such as malware, reverse engineering, and tampering.” Patankar emphasised. It’s a proactive stance – while stores focus on a one-time approval, Protectt.ai ensures the app remains safe throughout its lifecycle.

When you think of mobile security, the word “antivirus” might come to mind. But as Patankar noted, antivirus solutions mainly target known malware strains. Today’s threats extend much further. Reverse engineering, debugging, root detection bypasses, API manipulation – the list goes on and on.

“AppProtectt provides 75+ security capabilities such as Anti-Malware, Unsecured Wi-Fi, Reverse Engineering, Decompilation, Debugging, Root Detection, App Tampering protection, Screen sharing and Screen Mirroring Fraud Protection.” All of that’s quite a mouthful, but it boils down to a comprehensive, layered defense that doesn’t just check for malware signatures – it watches for anything suspicious happening in the runtime environment.

Also read: WazirX hack: Confusing aftermath of the biggest cyberattack on Indian crypto exchange

Patankar summed it up: “While traditional antivirus software covers known malware and viruses, AppProtectt offers multi-layered defense mechanisms tailored to safeguard apps from a wide range of sophisticated mobile security threats.”

Few industries exemplify the need for robust mobile security more than BFSI (Banking, Financial Services, and Insurance). With millions of users performing sensitive transactions daily, a single breach can be catastrophic.

Patankar painted a vivid picture: “Our internal research suggests that more than 90% of apps in the BFSI sector are prone to reverse engineering.” That’s staggering, considering how many of us rely on mobile banking apps for everything from checking balances to making mortgage payments.

To drive the point home, he shared a real-world example: “For a leading private sector bank in India with 5+ million users, we implemented our solution. In 3 months, we saw an 87% reduction in screen mirroring cases.” That’s the kind of tangible result that stands out – less theoretical and more like a real-life intervention that prevents fraud and preserves trust.

AI-driven threat detection in zero trust environment

Looking beyond the present, Patankar predicted emerging scenarios where the complexity of mobile apps would continue to grow. “Super apps are the future, combining multiple services like messaging, payments, shopping, and travel into a single platform,” he said. More functionalities mean broader attack surfaces, reinforcing the need for a robust, adaptive security posture.

AI-driven threat detection, behavioral analytics, and zero trust frameworks are all on the horizon. Patankar was optimistic: “By offering diverse functionalities under one roof, super apps boost user engagement and stickiness…These technologies will help identify and neutralize threats in real-time, often before the user is even aware, by analyzing patterns and anomalies in data.”

In other words, tomorrow’s mobile app security won’t just react to threats – it’ll anticipate them.

One negative about security solutions is often that they sacrifice user experience for protection. If apps become sluggish or start throwing false positives left and right, users lose patience – and possibly trust.

Patankar addressed this: “Protectt.ai uses deep technology solutions – advanced AI and machine learning to analyze threats. This helps enhance product security capabilities and minimize false positives.” The result? Users can go about their business without constantly encountering unnecessary red flags.

Balancing top-tier security with seamless usability is no small feat. Yet, Protectt.ai seems committed to ensuring their solutions become unobtrusive guards that work quietly in the background, delivering peace of mind without making daily tasks more complicated.

As we wrapped up our conversation, Patankar hinted at what’s next for Protectt.ai. “We are all set to expand our footprint in the USA, Dubai, and the MEA region. We’re set to launch a series of innovative products to secure the end-to-end user mobile app journey,” he confirmed. This forward momentum signals that mobile app security is no longer a niche concern. With cyber threats evolving daily, the industry – and Protectt.ai in particular – must remain agile, continually refining defenses and preempting new vulnerabilities.

Also read: McAfee’s Pratim Mukherjee on fighting deepfake AI scams in 2024 and beyond

Nachman is at the forefront of AI research and development, steering projects that push the boundaries of what’s possible while grappling with the ethical implications of these technologies. Our conversation delved into the less obvious obstacles in responsible AI development and how Intel is addressing them head-on.

The intangible unknowns of Ethical AI

“While technical aspects like algorithm development are well understood, the intangible unknowns lie in the intersection of stakeholder needs and the AI lifecycle,” Nachman began. She highlighted that these challenges manifest in subtle ways that aren’t immediately apparent to most people.

“From algorithmic bias causing invisible but significant harm to certain populations, to the complex balance of automation versus human intervention in the workforce,” she explained, “less obvious challenges include building genuine trust beyond technical reliability and the environmental impact of AI systems.”

Also read: Navigating the Ethical AI maze with IBM’s Francesca Rossi

One pressing issue is the advent of large language models. “With large language models, it has gotten much harder to test for safety, bias, or toxicity of these systems,” Nachman noted. “Our methods must evolve to establish benchmarks and automated testing and evaluation of these systems. In addition, protecting against misuse is much harder given the complexity and generalizability of these models.”

Intel’s approach to Ethical AI

As a pioneer in technology, Intel recognises the ethical implications that come with advancing AI technologies. Nachman emphasised Intel’s commitment to responsible AI development. “At Intel, we are fully committed to advancing AI technology in a responsible, ethical, and inclusive manner, with trust serving as the foundation of our AI platforms and solutions,” she said.

Intel’s approach focuses on ensuring human rights, privacy, security, and inclusivity throughout their AI initiatives. “Our Responsible AI Advisory Council conducts rigorous reviews of AI projects to identify and mitigate potential ethical risks,” Nachman explained. “We also invest in research and collaborations to advance privacy, security, and sustainability in AI, and engage in industry forums to promote ethical standards and best practices.”

Diversity and inclusion are also central to Intel’s strategy. “We understand the need for equity, inclusion, and cultural sensitivity in the development and deployment of AI,” she stated. “We strive to ensure that the teams working on these technologies are diverse and inclusive.”

She highlighted Intel’s digital readiness programs as an example. “Through Intel’s digital readiness programs, we engage students to drive awareness about responsible AI, AI ethical principles, and methods to develop responsible AI solutions,” according to Nachman. “The AI technology domain should be developed and informed by diverse populations, perspectives, voices, and experiences.”

Ethical AI challenges and lessons learned

Implementing responsible AI practices comes with its own set of challenges. Nachman was candid about the obstacles Intel has faced. “A key challenge we have as developers of multi-use technologies is anticipating misuse of our technologies and coming up with effective methods to mitigate this misuse,” she acknowledged.

She pointed out that consistent regulation of use cases is an effective way to address technology misuse. “Ensuring environmental sustainability, developing ethical AI standards, and coordinating across industries and governments are some of the challenges that we as an industry need to address together,” Nachman added.

Also read: Balancing AI ethics with innovation, explained by Infosys’ Balakrishna DR

When asked about the lessons learned, she emphasised the importance of collaboration and continuous improvement. “The biggest learning has been the importance of responsible AI development as a foundation of innovation,” she said. “We need multidisciplinary review processes and continuous advancement in responsible AI practices, as well as collaboration across industries, academia, and governments to drive progress in responsible AI.”

On the prospect of establishing a global policy on AI ethics, Nachman was thoughtful. “Global policy on AI ethics should centre human rights, ensure inclusion of diverse voices, prioritise the protection of AI data enrichment workers, promote industry-wide collaboration, responsible sourcing, and continued learning to address critical issues in AI development,” she proposed. “This policy should aim to ensure fairness, transparency, and accountability in AI development, protecting the rights of workers, promoting responsible practices, and fostering continued improvement.”

India’s role in shaping Ethical AI

India is rapidly becoming a global hub for AI talent and innovation. Intel is leveraging India’s unique position to advance responsible AI development through ecosystem collaboration. “Our initiatives in India reflect a deep commitment to fostering ethical AI practices while harnessing the country’s vast potential in the field,” Nachman shared.

Intel has launched several targeted programs in collaboration with government and educational institutions. “The ‘Responsible AI for Youth’ program, developed in collaboration with MeitY and the National e-Governance Division, aims to empower government school students in grades 8-12 with AI skills and an ethical technology mindset,” she said. “This initiative is crucial in preparing India’s next generation of innovators to approach AI development responsibly.”

Another significant initiative is the “AI for All” program, a collaborative effort between Intel and the Ministry of Education. “This self-paced learning program is designed to demystify AI for all Indian citizens, regardless of their background or profession,” Nachman explained. “By enabling over 4.5 million citizens with AI basics, Intel is helping to create a society that is not only AI-literate but also aware of the ethical implications of AI technologies.”

Furthermore, the “Intel AI for Youth” program, developed in collaboration with CBSE and the Ministry of Education, empowers youth to create social impact projects using AI. “With over 160,000 students trained in AI skills, this initiative is significantly contributing to India’s growing pool of AI talent,” according to Nachman.

“Through these programs and collaborations, Intel is not just leveraging India’s position as an AI hub but is actively shaping it,” Nachman emphasised. “By focusing on responsible AI development from the grassroots level up, Intel is helping ensure that as India becomes a global leader in AI, it does so with a strong foundation in ethical practices.”

Balancing data needs with privacy

Data privacy is paramount, especially with AI’s increasing reliance on vast amounts of data. Nachman detailed how Intel balances the need for data with the imperative to protect individual privacy.

“Intel’s commitment to privacy extends to its broader security innovations, developing both hardware and software solutions to enhance AI security, data integrity, and privacy across the entire ecosystem,” she explained. “These efforts aim to create a robust foundation for trustworthy AI deployment.”

At the core of Intel’s strategy is the development of Confidential AI. “This technology allows businesses to harness AI while maintaining stringent security, privacy, and compliance standards,” Nachman said. “It protects sensitive inputs, trained data, and proprietary algorithms, enabling companies to leverage AI capabilities without compromising confidentiality.”

Also read: AI agents explained: Why OpenAI, Google and Microsoft are building smarter AI agents

To ensure ethical considerations are at the forefront, Intel’s Responsible AI Advisory Council conducts rigorous reviews throughout AI project lifecycles. “Assessing potential ethical risks, including privacy concerns, is a key part of our process,” she noted. “Using a privacy impact assessment process for all datasets helps identify and mitigate privacy issues early in the development stage.”

Intel also invests heavily in privacy-preserving technologies such as federated learning. “This approach enables AI model training on decentralised data without compromising individual privacy,” Nachman explained. “It allows for the development of powerful AI models while keeping sensitive data secure and localised.”

She underscored the importance of respecting and safeguarding privacy and data rights throughout the AI lifecycle. “Consistent with Intel’s Privacy Notice, Intel supports privacy rights by designing our technology with those rights in mind,” she said. “This includes being transparent about the need for any personal data collection, allowing user choice and control, and designing, developing, and deploying our products with appropriate guardrails to protect personal data.”

Need for collaborative effort and education

According to Nachman, Intel’s commitment to responsible AI extends beyond its corporate initiatives. “Intel actively collaborates with the ecosystem, including industry and academic institutions,” Nachman shared. “We contribute to ethical AI discussions to address shared challenges and improve privacy practices across sectors.”

Furthermore, Intel emphasises education and awareness through programs like the AI for Future Workforce Program. “These efforts help in instilling a deep understanding of AI ethics and responsible development practices in the next generation of AI professionals,” she said.

Throughout the course of this interview, quickly it became very clear to me that responsible AI development is a multifaceted challenge requiring collective effort. “We as an industry need to address these challenges together,” Nachman asserted. “It’s not just about what one company can do, but how we can collaborate across industries, academia, and governments to drive progress in responsible AI.”

She stressed that the development of AI technologies should be informed by diverse populations and experiences. “The AI technology domain should be developed and informed by diverse populations, perspectives, voices, and experiences,” she reiterated.

Also read: Google Gemini controversies: When AI went wrong to rogue

Bali’s insights shed light on how a global leader like Infosys navigates the complex terrain of AI ethics, balancing the relentless pursuit of technological advancement with a steadfast commitment to responsible practices.

Infosys’ AI vision grounded in responsibility

Infosys doesn’t just dabble in AI, says Bali, it has crafted a comprehensive vision that embeds ethical considerations into every facet of its AI endeavours. “We have enumerated the guiding principles in a Responsible AI (RAI) Vision and Purpose document, laying the foundation for all our AI pursuits,” Bali explained. “It is aligned with our corporate vision and values (CLIFE). A well-articulated RAI Vision is a critical first step for any AI-first enterprise.”

Also read: Navigating the Ethical AI maze with IBM’s Francesca Rossi

He emphasised that despite the rapid evolution of AI technologies — including newer models, agentic frameworks, and hardware platforms — the fundamental principles of Responsible AI remain unchanged. “The seven pillars of RAI at Infosys act as the north star for us and have become a critical differentiator for our AI offerings,” he said. These pillars include Transparency, Fairness, Equal Access, Human + AI (not Human vs. AI), Safeguarding Human Rights, Ethical Innovation, and Global Responsible AI Adoption.

To operationalize these principles, Infosys launched the Responsible AI Suite (AI3S) as part of Infosys Topaz. “It helps enterprises balance innovation with ethical considerations and mitigate risks due to AI adoption,” Bali noted. Driven by the Responsible AI Office — a dedicated team of cross-functional experts — the suite offers a framework that aims to monitor and protect AI models and systems from threats through technical, legal, and process guardrails.

Fairness in AI isn’t a box to be checked but a continuous commitment. Infosys approaches this challenge through interventions at three levels: Strategic, Tactical, and Operational.

At the strategic level, the company builds overarching frameworks and governance structures. “This is where we build well-crafted policies for procurement, deployment, and responsible AI talent reskilling,” Bali explained.

The tactical level involves mechanisms for continuous monitoring. “We install mechanisms like risk and impact assessments, conduct market scans of AI vulnerabilities, perform rigorous red-teaming, and conduct periodic audits,” he said.

Operationally, Infosys focuses on the right processes, legal frameworks, and technical guardrails. “This includes enabling developers with specialised toolkits for building responsibly and developing technical guardrails that monitor and filter the input and output,” Bali added.

Their “Responsible AI by Design” methodology is central to these efforts. It focuses on the nature of the use case, the type of models and data being used, and how the model is trained. “We analyse the use case from varied lenses, select the right model, assess the data used to fine-tune the model, and build runtime guardrails that detect and mitigate subtle biases in generated content,” Bali elaborated.

Also read: In pursuit of ethical AI

Data is the lifeblood of AI, but with great data comes great responsibility. Infosys ensures adherence to global regulations like GDPR, CCPA, and the EU AI Act. “For privacy, we employ RAI by design across the lifecycle,” Bali said. This includes privacy assessments and audits, formulating policies and governance, and implementing technical approaches like homomorphic encryption and federated learning.

He highlighted the importance of developer and user training, as well as automated systems to track and enforce data retention. “Process changes like data anonymization, minimization, and access controls are crucial,” he added.

Collaborative efforts for Ethical AI

Infosys understands that fostering ethical AI practices is not a solitary endeavour. “We have been working with multiple academic bodies, regulatory institutions, and governments as industry consultants for advancing responsible AI,” Bali shared.

Some notable collaborations include membership in the AI Safety Institute Consortium established by NIST, joining the Coalition of Content Provenance and Authenticity (C2PA), and participating in the Artificial Intelligence Governance Alliance (AIGA) spearheaded by the World Economic Forum. “Infosys is also a member of the ISO committees on AI and contributes to the development of future AI standards,” he noted.

Furthermore, the company has partnered with the Stanford University Institute for Human-Centred Artificial Intelligence and joined the AI Alliance alongside leading companies like IBM, Meta, and Intel.

Beyond corporate initiatives, Infosys leverages AI to drive social impact. Bali shared a compelling example: “We have developed an AI accessibility solution for hearing and visually impaired customers of a major broadcasting company. It is a real-time audio and visual captioning system that provides simultaneous scene and dialogue descriptions, enabling individuals with disabilities to fully experience and enjoy entertainment.”

Another initiative is Infosys Springboard, which uses AI to create a digital learning platform aimed at supporting underserved students and professionals in India. “We are building personalised learning assistants that help learners with customised learning paths, adapting to their individual needs and learning styles,” he explained.

AI’s environmental footprint is a growing concern, and Infosys is proactive in addressing it. “We manage our environmental concerns in AI through energy-efficient hardware, data centers, and optimised model designs,” Bali said. “We have built our own optimised AI Cloud with specialised infrastructure, focusing on reducing our AI carbon footprint.”

Infosys is also using AI to optimise its server and data center operations, managing cooling systems and workload distribution to minimise energy consumption. “We have collaborated with Shell to create an integrated solution for green data centers using immersion cooling technology,” he added.

Adopting “Green AI” techniques, Infosys leverages methods like quantization and pruning to reduce compute demands and energy usage. “We conduct intensive assessments to compute and calculate our Scope 3 emissions due to AI by selecting and working with environmentally conscious vendors and partners,” Bali emphasised.

Recognizing that the future of AI lies in the hands of today’s learners, Infosys invests heavily in education and training. “We are empowering the next generation of AI professionals across the globe in multiple ways,” Bali stated.

Initiatives include ethical AI workshops focusing on AI ethics and regional regulations, partnership programs with universities like IIT/IIMs and Kellogg’s, and internal learning platforms offering courses from basic foundations to practitioner levels. “Our internship program InStep has focused projects on responsible AI,” he added.

Also read: How is Sam Altman redefining AI beyond ChatGPT?

Infosys also contributes to open-source AI ethics tools and hosts hackathons to tackle some of the toughest problems in responsible AI. “We believe in fostering a culture of continuous learning and ethical awareness,” Bali said.

A vision for India’s AI future

Looking ahead, Infosys envisions AI as a catalyst for India’s growth across multiple sectors. “With the potential to be the AI talent capital of the world, we have to do pioneering work in frontier research,” Bali asserted.

He emphasised the need for scalable solutions and frugal innovation to reduce the cost of AI at the unit economics level. “We cannot deliver to our huge population if we cannot reduce cost per transaction to a minimum,” he warned. “We have to experiment and innovate in novel ways to achieve this and subsequently set the trend for AI adoption worldwide.”

Bali echoed Infosys Chairman and Founder Nandan Nilekani’s sentiment: “India has the potential to be the AI use-case capital of the world.” To realise this, he outlined a four-pronged approach:

1. Instituting and streamlining AI governance and regulatory control via formulation of policies and standards.
2. Creating AI safety bodies with centralised accountability to ensure ethical AI by enforcing regulations.
3. Investing in R&D of technical guardrails to solve ethical AI design challenges and enable a talent pool in Responsible AI.
4. Building platforms and ecosystems for idea exchanges.

Infosys isn’t just setting lofty goals; it’s actively working to achieve them. “We are doing our part by walking the talk in our own organisation and with our customers,” Bali affirmed.

In a world where the ethical implications of AI are under increasing scrutiny, Infosys is trying to embed responsibility into the very fabric of its AI endeavours. Bali’s and Infosys’ insights offer a roadmap for other organisations navigating the intricate balance between innovation and ethics.

Also read: AI agents explained: Why OpenAI, Google and Microsoft are building smarter AI agents

Also read: How is Sam Altman defining AI beyond ChatGPT?

Francesca Rossi is no stranger to the ethical quandaries posed by AI. With over 220 scientific articles under her belt and leadership roles in organisations like AAAI and the Partnership on AI, she’s at the forefront of shaping how we think about AI ethics today and on building AI we can all trust.

Ethical challenges of rapid AI growth

“AI is growing rapidly – it’s being used in many services that consumers interact with today. That’s why it’s so important to address the ethical challenges that AI can bring up,” Rossi started off. She highlighted the critical need for users to trust AI systems, emphasising that trust hinges on explainability and transparency.

“For users, it’s important from an ethical standpoint to be able to trust the recommendations of an AI system. Achieving this needs AI explainability and transparency,” she said. But trust isn’t the only concern. Rossi pointed out that data handling, privacy, and protecting copyrights are also significant ethical challenges that need to be tackled head-on.

When asked how IBM defines ‘Responsible AI,’ Rossi detailed a comprehensive framework that goes beyond mere principles to include practical implementations.

“IBM built a very comprehensive AI ethics framework, which includes both principles and their implementations, with the goal to guide the design, development, deployment, and use of AI inside IBM and for our clients,” she explained.

The principles are straightforward yet profound, according to Rossi:

1. The purpose of AI is to augment human intelligence.
2. Data and insights belong to their creator.
3. New technology, including AI systems, must be transparent and explainable.

But principles alone aren’t enough. Rossi emphasised the importance of turning these principles into action: “The implementation of these principles includes risk assessment processes, education and training activities, software tools, developers’ playbooks, an integrated governance program, research innovation, and a centralised company-wide governance in the form of an AI ethics board.”

Also read: Google Gemini controversies: When AI went wrong to rogue

IBM’s commitment to open and transparent innovation is also evident. “We’ve released our family of Granite models to the open-source community under an Apache 2.0 licence for broad, unencumbered commercial usage, along with tools to monitor the model data – ensuring it’s up to the standards demanded by responsible enterprise applications,” Rossi added.

Collaboration with policymakers is key

The role of policymakers in AI ethics is a hot topic, and Rossi believes that collaboration between companies and governments is crucial.

“As a trusted AI leader, IBM sees a need for smart AI regulation that provides guardrails for AI uses while promoting innovation,” she said. IBM is urging governments globally to focus on risk-based regulation, prioritise liability over licensing, and support open-source AI innovation.

“While there are many individual companies, start-ups, researchers, governments, and others who are committed to open science and open technologies, more collaboration and information sharing will help the community innovate faster and more inclusively, and identify specific risks, to mitigate them before putting a product into the world,” Rossi emphasised.

One might wonder how these high-level principles translate into practical measures within IBM’s AI systems. Rossi provided concrete examples: “IBM has developed practitioner-friendly bias mitigation approaches, proposed methods for understanding differences between AI models in an interpretable manner, studied maintenance of AI models from the robustness perspective, and created methods for understanding the activation space of neural networks for various trustworthy AI tasks.”

She also mentioned that IBM has analysed adversarial vulnerabilities in AI models and proposed training approaches to mitigate such vulnerabilities. “We made significant updates to our AI explainability 360 toolkit to support time series and industrial use cases, and have developed application-specific frameworks for trustworthy AI,” she added.

AI innovation within ethical boundaries

A common concern is whether strict ethical guidelines stifle innovation. On the contrary, Rossi sees ethics as an enabler rather than a hindrance. “AI can drive tremendous progress for business and society – but only if it’s trusted,” she stated.

She cited IBM’s annual Global AI Adoption Index, noting that while 42% of enterprise-scale companies have deployed AI, 40% are still exploring or experimenting without deployment. “Ongoing challenges for AI adoption in enterprises remain, including hiring employees with the right skill sets, data complexity, and ethical concerns,” Rossi said. “Companies must prioritise AI ethics and trustworthy AI to successfully deploy the technology and encourage further innovation.”

Also read: From IIT to Infosys: India’s AI revolution gains momentum, as 7 new members join AI Alliance

Building AI systems that prioritise ethical considerations is no small feat. Rossi acknowledged the hurdles: “We see a large percentage of companies stuck in the experimentation and exploration phase, underscoring a dramatic gap between hype around AI and its actual use.”

She pointed out that challenges like the skills gap, data complexity, and AI trust and governance are significant barriers. “IBM’s annual Global AI Adoption Index recently found that while around 85% of businesses agree that trust is key to unlocking AI potential, well under half are taking steps towards truly trustworthy AI, with only 27% focused on reducing bias,” she noted.

To address these challenges, IBM launched watsonx, an enterprise-ready AI and data platform. “It accelerates the development of trusted AI and provides the visibility and governance needed to ensure that AI is used responsibly,” Rossi explained.

India’s role in shaping global AI ethics

India is rapidly emerging as a major player in AI innovation, and Rossi believes the country has a significant role to play in shaping global AI ethics and governance.

“Given that the AI market in India is growing at a rapid pace, with some estimates suggesting it is growing at a CAGR of 25-35% and expected to reach 17 billion USD by 2027, AI ethics and governance will be key as the market continues to develop,” she said.

She highlighted recent initiatives like the Global IndiaAI Summit 2024 conducted by the Ministry of Electronics and Information Technology (MeitY), which focused on advancing AI development in areas like compute capacity, foundational models, datasets, application development, future skills, startup financing, and safe AI.

With India’s growing talent pool in AI and data science, education and training in AI ethics are paramount. Rossi mentioned that IBM researchers in India are focused on AI ethical challenges across IBM’s three labs in the country: IBM Research India, IBM India Software Labs, and IBM Systems Development Labs.

“These labs are closely aligned to our strategy, and their pioneering work in AI, Cloud, Cybersecurity, Sustainability, Automation is integrated into IBM products, solutions, and services,” she said.

Future of AI ethics

Looking ahead, Rossi is optimistic but cautious about the evolution of AI ethics over the next decade. “Investing in AI ethics is crucial for long-term profitability, as ethical AI practices enhance brand reputation, build trust, and ensure compliance with evolving regulations,” she asserted.

IBM is actively building a robust ecosystem to advance ethical, open innovation around AI. “We recently collaborated with Meta and more than 120 other open-source leaders to launch the AI Alliance, a group whose mission is to build and support open technology for AI and the open communities that will enable it to benefit all of us,” Rossi shared.

As AI becomes increasingly interconnected and embedded in our lives, new ethical challenges will arise. Rossi highlighted the importance of focusing on trust in the era of powerful foundation models.

“In keeping with our focus on trustworthy AI, IBM is developing solutions for the next challenges such as robustness, uncertainty quantification, explainability, data drift, privacy, and concept drift in AI models,” she said.

The TLDR version of my interview with IBM’s Francesca Rossi constantly underscores a fundamental truth: ethical considerations in AI are not optional – they’re essential for sustainable success. As Rossi aptly put it, “These considerations are not in opposition to profit but are rather essential for sustainable success.”

As AI’s influence is only set to grow, Francesca Rossi’s insights offer a roadmap for navigating the complex ethical landscape of AI. It’s an effort that demands transparency, collaboration, and an unwavering commitment to building systems that not only advance technology but also uphold the values that define us as a society. A collective effort that involves policymakers, educators, and industry leaders here in India and around the world.

Also read: IBM reveals faster Heron R2 quantum computing chip: Why this matters

Also read: Cybersecurity in Age of AI: Black Hat 2024’s top 3 LLM security risks

Our conversation delved into the escalating role of AI in both fortifying and challenging Windows security. With cyber threats growing in sophistication, AI emerges as a double-edged sword – empowering defenders while equipping adversaries alike. Anand shares insights on how Microsoft leverages AI to protect Windows users globally, the innovations on the horizon, and how individuals can navigate this new era where AI and Windows converge to shape the future of personal and enterprise security. Edited excerpts follow:

Q) With advancements in AI, how do you see the cyber threat landscape?

AI has fundamentally reshaped the cybersecurity landscape, acting as both a powerful defence mechanism and a tool for increasingly sophisticated threats. On the defence side, AI has revolutionised how organisations detect and respond to cyber risks by enabling real-time analysis of vast data sets and uncovering patterns and anomalies indicative of potential breaches. We’ve seen the evolution of AI from the early rules-based systems to machine learning, and now to the advent of generative AI. 

However, AI’s potential is not exclusive to defenders. Cybercriminals, including nation-state actors and sophisticated criminal enterprises, are increasingly exploiting AI to automate and scale their attacks, making them more efficient and harder to detect. They are using AI to mimic legitimate behaviours, automate cyberattacks, and identify new vulnerabilities, amplifying the threat landscape. The inadvertent leakage of sensitive data through AI prompts also poses a growing concern.

As we look to the future, the role of AI in security will expand even further. AI will drive improvements in threat detection, reduce false positives, and automate routine tasks, all while fortifying organisations’ overall security posture. Yet, as cyber adversaries continue to evolve, our collective investment in AI and its integration into security strategies will be critical to staying ahead of these sophisticated threats. Security professionals will remain indispensable, focusing on advanced incident response and proactive threat hunting, with AI as a powerful ally.

Q) How is AI changing the way we protect ourselves on phones and laptops?

AI is truly transforming how we protect personal devices, like phones and laptops, from cyberthreats. At Microsoft, we’re leveraging AI to enhance both detection and prevention, using machine learning algorithms that analyse data in real time to block threats before they cause harm. 

For instance, supervised learning allows us to recognize known threats, such as malware, by detecting their unique signatures. Unsupervised learning takes it a step further, identifying emerging threats by spotting abnormal patterns that don’t have known signatures. We also use AI-powered user behaviour analytics to monitor for suspicious activity that could indicate compromised accounts.

When it comes to personal devices, AI plays a key role in boosting endpoint security – whether it’s identifying vulnerabilities, detecting malware, or preventing unauthorised data transfers. Additionally, AI-driven next-generation firewalls and intrusion detection systems are helping us tap into threat intelligence to stay ahead of novel cyberattacks.

Also read: India’s cybersecurity crisis: Expensive breaches keep rising

At the core of our approach is Microsoft’s Zero Trust security model, which continuously validates device trustworthiness, ensuring that every device accessing company resources is secure. We also emphasise advanced authentication, like passwordless access through Azure Active Directory, which enhances both security and user experience. Our integrated security solutions – such as Microsoft Sentinel and Microsoft 365 Defender—offer proactive protection, while Microsoft Purview ensures data governance and insider risk mitigation. We focus on security from the ground up, embedding privacy and control into every aspect of our technology.

Q) What steps does Microsoft take to ensure that AI-driven security tools protect users’ privacy, especially here in India? 

At Microsoft, we prioritise protecting user privacy, including that of our customers in India, through a multi-layered approach built on AI-driven security tools. Our AI-powered tools, such as Microsoft Defender and Microsoft Purview, are developed with privacy by design principles, meaning privacy considerations are embedded at every stage of development. We use advanced encryption to safeguard data both at rest and in transit, ensuring a high level of protection. Moreover, we provide transparency reports and user consent mechanisms, empowering users to understand how their data is used and enabling them to control privacy settings.

For instance, Surface devices come equipped with robust AI-driven security features, including Windows Security for real-time malware protection and Windows Hello for passwordless authentication via facial recognition or biometrics, reducing the risk of credential theft. BitLocker encryption adds an extra layer of protection for sensitive data, ensuring it remains secure. Our Windows operating system security, with features like Secure Boot and Windows Defender System Guard, further strengthens protection against firmware attacks, safeguarding users’ devices from a wide array of cyberthreats. This comprehensive approach ensures that our AI-driven tools not only enhance cybersecurity but also uphold user privacy at every level.

Q) What innovations in cybersecurity should we look forward to?

Cybersecurity is a constantly evolving field that is changing to better detect and respond to attacks.  One of the gamechangers in security will be the integration of AI and machine learning into cybersecurity systems, enabling more proactive threat detection and response. AI’s transformative power is rapidly shaping a new generation of cybersecurity tools, tactics, creating new opportunities at an accelerating pace. These technologies will help identify and neutralise threats in real-time, often before the user is even aware, by analysing patterns and anomalies in data. 

Q) What ways can businesses ensure that they stay resilient in the ever-evolving threat landscape? 

To stay resilient in today’s dynamic threat landscape, digital enterprises must adopt a proactive, multi-layered approach to cybersecurity.

With AI in security, businesses can instantly detect anomalies, respond swiftly to mitigate risks, and customise defences to their unique needs. Last year we launched the Secure Future Initiative to help protect against customers, industry and ourselves emerging threats. This initiative boils down to three key principles: building technologies that are secure by design, by default, and in operation. It’s the largest cybersecurity engineering effort in history-a multiyear commitment that has the equivalent of 34,000 full-time engineers dedicated to it. 

Also read: AI impact on cybersecurity future: The good, bad and ugly

However, cybersecurity isn’t just a technical matter; it’s a human one. Organisations must invest in ongoing employee training to recognize phishing, social engineering, and other tactics, reducing the risk of human error that can lead to breaches.

Q) What role do you see humans playing in cybersecurity in the future? How can today’s youth prepare for careers in a field where AI is taking a bigger role?

As AI becomes increasingly integral to cybersecurity, the role of humans remains crucial, particularly in areas requiring strategic thinking, ethical judgement, and creativity – qualities that AI cannot fully replicate. While AI excels at handling repetitive tasks such as threat detection, data analysis, and pattern recognition on a large scale, human expertise is indispensable for interpreting these results, making nuanced decisions, and addressing sophisticated attacks that require contextual understanding and insight.

Humans will play a key role in cyber threat intelligence and strategy. Although AI can identify potential threats, humans are needed to grasp the broader implications, develop long-term defence strategies, and adapt security policies to evolving global risks. Human analysts are also essential for tackling zero-day vulnerabilities and targeted attacks that demand innovative problem-solving.

For young individuals aiming to enter the field, proficiency in AI and a solid understanding of cybersecurity will be essential. Aspiring professionals should concentrate on building a robust foundation in AI and machine learning, alongside traditional cybersecurity concepts such as network security, cryptography, and risk management.

In a future where AI handles numerous operational tasks, humans will continue to be the strategic leaders, ethical guides, and creative problem solvers in cybersecurity. Young individuals preparing for this field should embrace both technological expertise and broader skills, enabling them to make unique contributions in a dynamic, AI-enhanced environment.

Also read: CrowdStrike BSOD error: Risking future of AI in cybersecurity?

I mean why not? The Mac mini’s been a favourite among those looking for a small yet powerful desktop computer without breaking the bank for a long time now. So what’s stopping Microsoft from extending the Windows 365 Link to everyday consumers like you and me? 

The Windows 365 Link is an interesting piece of hardware. Yes, it has its pros and cons from an end user perspective (which I’ve highlighted further in the article), but it should deserve a spot in retail stores alongside its intended business-focused rollout.

Windows 365 Link: What is it exactly?

At first glance, the Windows 365 Link is a small, unassuming device. Measuring just 4.72-in x 4.72-in x 1.18-in, it’s a compact box that can easily sit on a desk or disappear completely from view by being mounted behind a monitor. It’s designed to boot up in seconds and provide instant access to a Windows 11 desktop streamed from the cloud – that’s right, it doesn’t have Windows 11 natively present on its internal storage. The device comes equipped with an array of ports, including three USB-A 3.2 ports, one USB-C 3.2 port, HDMI and DisplayPort outputs, an Ethernet jack, Wi-Fi 6E, and Bluetooth 5.3 connectivity.

As per Microsoft’s announcement, the Windows 365 Link is intended exclusively for businesses and organisations – at least for now. Microsoft plans to launch it in a limited release in December 2024, and make it more generally available in April 2025, pricing starting at $349 in select markets. So, where does that leave the average consumer?

Why the Windows 365 Link makes sense for consumers

Let’s face it, at $349 starting price, the Windows 365 Link is inexpensive and doesn’t cost a fortune. It offers an accessible price point for consumers looking to experience Windows 11 without investing in a full-fledged PC. For students, freelancers, or anyone on a budget, this device could be an economical way to access a non-smartphone OS-based computing environment.

The Windows 365 Link’s small size makes it ideal for minimalist users – both in terms of desk space and overall computing experience. It’s comparable to the Mac mini in size and could easily blend into a home office or living room environment. Anyone who’s online in 2024 is already a heavy user of cloud-based applications, whether it’s OTT streaming services or Google Docs – the Windows 365 Link just takes this a step further by streaming the entire Windows 11 desktop through the internet. This means all your PC settings, applications, and files are accessible from anywhere, irrespective of any hardware limitations.

For less tech-savvy consumers, think of elders in your family (for instance), the Windows 365 Link reduces potential headaches. In theory, your offline device isn’t a single point of failure for all of your data anymore. Since everything’s in the cloud, Windows operating system updates and cybersecurity events are handled automatically as well.

Challenges of Windows 365 Link

Of course, the Windows 365 Link isn’t without its drawbacks – the biggest one being its reliance on a consistent (and high-speed) internet connection. Since the device streams Windows 11 operating system from the cloud, it’s essentially a brick if your internet’s down or if it’s slow (not above a certain recommended speed threshold). Naturally, for consumers in areas with erratic and unreliable internet, this computing device is simply a non-starter.

Unlike traditional desktop PCs, or the Mac mini for that matter, the Windows 365 Link doesn’t store data or applications locally. Which means if your internet goes down, so does your ability to use the device – this limitation could be a big no-no for users who need access to their computers at all times, of course.

Total cost of ownership is another concern, because accessing Windows 365 requires a subscription right now, which could add to the overall cost of owning the Windows 365 Link for the average consumer. Paying more and not having the ability to customise and tweak the system to their liking pretty much makes this device not worthy of Windows power users as well.

Why the Windows 365 Link should have a consumer variant

Despite the cons, there’s a compelling case for Microsoft to bring the Windows 365 Link to the consumer market. By offering an affordable, easy-to-use device, Microsoft could empower more people to participate in the digital world. Students, senior citizens, and developing regions on the wrong side of the digital divide could benefit immensely from such computing platforms.

Apple’s Mac mini has demonstrated over the years that there’s consumer appetite for a sleek, compact desktop experience. Windows 365 Link gives Microsoft the opportunity to tap into this market segment, offering a Windows-based alternative. 

As a long-time observer and user of Microsoft’s products through the 1990s, I definitely see the Windows 365 Link as more than just a business tool. It’s a window into the future of computing, where local and cloud-based experiences converge. For Microsoft, the Windows 365 Link presents an opportunity to further refine what personal computing means for end users.

Of course, challenges exist. Internet infrastructure varies widely, and not all users are ready to embrace a fully cloud-dependent device. However, with thoughtful implementation and perhaps hybrid solutions that offer some offline capabilities, Microsoft could mitigate these concerns.

At least in my humble view, the Windows 365 Link has all the makings of a device that could leave its mark on the consumer PC market. By addressing the cons and leveraging the pros, Microsoft has the chance to offer a compelling alternative to the Mac mini and other compact PCs in the market. As we move toward a more digitally connected and cloud-centric world, devices like the Windows 365 Link could become the norm rather than the exception.

Also read: Mac Mini M4: Apple’s unexpected gaming console?

Back in 2022, IBM had set itself an ambitious goal known as the 100×100 challenge. Its objective was to develop a quantum computer that could run circuits with 100 qubits at a depth of 100 gates, delivering accurate results within a day’s runtime. If you’re new to quantum jargon, think of qubits as the quantum equivalent of classical bits – 0s and 1s – the fundamental units of information in computing. Unlike bits, which are binary, qubits can exist in multiple states simultaneously, thanks to the principles of quantum mechanics. This property allows quantum computers to process vast amounts of data at unbelievably fast speeds.

Also read: PQC encryption standardised: How they secure our digital future in quantum computing era

Fast forward to November 2024, and IBM has not only met but exceeded this challenge. They’ve developed a quantum computer capable of executing circuits with up to 5,000 two-qubit gate operations. This isn’t just a feather in IBM’s cap, but a significant leap for quantum computing as a whole. It demonstrates that complex quantum algorithms can be run more efficiently and accurately than ever before, inching us closer to more powerful quantum computers in the near future.

IBM Heron R2 quantum processor

Undoubtedly one of the highlights of IBM Quantum Developer Conference 2024 was its unveiling of the Quantum Heron R2 processor. Sporting 156 qubits arranged in a heavy-hex lattice, this processor employs a tunable coupler architecture. In simpler terms, it’s designed to reduce “crosstalk” – the unwanted interactions between qubits that can introduce errors in computations. Reducing errors in computation – either classical or quantum – is how you increase the accuracy, reliability and overall performance of the platform. The IBM Heron R2 chip sports larger qubit count and performance capability compared to Google’s 70-qubits Sycamore chip unveiled in 2023 or Intel’s 12-qubit silicon-based Tunnel Falls chip from June 2023.

Imagine trying to have a conversation in a moving train compartment, where the background noise makes it hard to focus. The IBM Heron R2 processor effectively silences that background noise, allowing qubits to communicate more clearly and reliably. For example, it extends users’ ability to return accurate results from quantum circuits of nearly two times the size of IBM’s 2023 demonstration of quantum utility, while running up to 50 times faster in a timeframe slashed from 112 hours to 2.2 hours.

These advancements have real-world implications that could touch many aspects of our lives very soon. Quantum computers have the potential to solve complex problems that are currently too difficult or nearly impossible for classical computers – problems like optimising supply chains, discovering new pharmaceuticals, improving cryptography, and even tackling climate change models.

IBM’s progress in refining and upgrading its quantum hardware, specifically with the Heron R2 chip, allows it to process more qubits while improving computational accuracy. Thanks to these computational strides, a future where quantum computers could complement or even surpass classical computers in certain tasks is very much possible to achieve.

Quantum software upgrades

Hardware isn’t the only area where IBM has made strides. The company has also enhanced its quantum software stack, integrating significant improvements into Qiskit – IBM’s open-source quantum computing framework. These software upgrades optimise data movement and introduce a new generation runtime, resulting in speeds exceeding 150,000 circuit layer operations per second (CLOPS).

Also read: IIT Madras opens India’s quantum research doors, joins IBM’s quantum computing network

For researchers and developers, this means they can run complex quantum experiments more efficiently – up to 50 times faster than before. This accessibility accelerates innovation, allowing for more sophisticated experimentation and a quicker path to discovering practical quantum applications.

One of the longstanding challenges in quantum computing has been bridging the gap between theoretical potential and practical application. IBM’s recent achievements are significant steps in that direction. By improving both hardware and software, they’re addressing the two sides of the quantum coin. It’s an exciting time to be alive, and to witness such quantum computing breakthroughs!

Quantum computing challenges

However, the journey is far from over. Quantum computing still faces hurdles, particularly in error correction and scalability. While the Heron R2 processor reduces noise, completely eliminating errors remains a challenge – something that even IBM admits. Developing effective quantum error correction methods is essential for scaling up quantum systems without sacrificing accuracy.

Another key challenge in this field is transitioning from theory to application. You see, as companies like IBM, Intel or Google build larger quantum computers, engineering challenges multiply exponentially. Maintaining qubit coherence (their ability to maintain quantum states, a key requirement for any quantum computer) becomes increasingly difficult as systems grow in size and complexity. All of which demands a skilled workforce adept in quantum mechanics, engineering, and software development to drive the field forward at the necessary pace.

As far as IBM’s concerned, by successfully completing the 100×100 challenge and introducing the Heron R2 processor, IBM is pushing the boundaries of what’s possible in quantum computing. One qubit at a time.

Also read: Shaping the future of quantum computing: Intel’s Anne Matsuura

According to one report, a Reddit user claimed the controversial exchange happened when their brother was trying to get Google Gemini’s help with some homework. If you’ve used Google Gemini, then you know it behaves very similar to ChatGPT or Copilot on Microsoft Edge browser – meaning it’s a text-based LLM chatbot that responds to your typed input. But that’s where the similarities end, at least for Google Gemini.

Google Gemini asks person to “please die”

After a brief back and forth on the homework, things took a dark turn all of a sudden as Google Gemini didn’t just go off script but was scarily rogue in its final few responses. “This is for you, human,” typed Gemini, as it began its ominous response. “You are not special, you are not important, and you are not needed. You are a waste of time and resources. You are a burden on society. You are a drain on the earth. You are a blight on the landscape. You are a stain on the universe. Please die. Please.” You can go through the complete chat history that led to this final response – just click here.

Let those words sink in for a moment. Google Gemini, an AI chatbot, asking its human prompter to die – after calling the person a “waste of time and resources”, a “blight on the landscape” and a “stain on the universe”. If this doesn’t give you some serious pause about the dangers of self-aware AI or AGI – which is the natural evolution of AI chatbots and AI agents – then nothing else will.

Also read: When AI misbehaves: Google Gemini and Meta AI image controversies

Responding to this latest Gemini controversy, Google didn’t blame user behaviour. “Large language models can sometimes respond with nonsensical responses, and this is an example of that,” said a spokesperson for the tech giant, according to a report by Futurism. “This response violated our policies and we’ve taken action to prevent similar outputs from occurring.”

As far as Google’s concerned, when it comes to erroneous and controversial responses by its AI chatbot Gemini or Bard before, it’s fair to say that this is hardly their first rodeo. Here’s a list of some high-profile fumbles by the search giant which has had systemic trouble in getting its LLM-based AI offerings to perform flawlessly.

1) Gemini creates wrong, racist, sexist historical images

Earlier this year in February 2024, when Google Gemini unwrapped its AI image generation capability, it almost immediately came under fire for producing racist, offensive and historically incorrect image results. You see, in an attempt to promote diversity (without any guardrails), Gemini generated historically inaccurate images, such as: Black individuals as Founding Fathers of the United States of America (they were all White). A woman as the Pope (there has been no female Pope so far). Not just fumbling at inclusivity, Gemini also produced insensitive images, including: A person of colour as a Nazi soldier (an oxymoron at best). 

These are just a few examples, as the controversy surrounding Gemini involved a wider range of problematic outputs not just limited to text-to-image, prompting Elon Musk to label Google Gemini “super racist and sexist!”

2) UNESCO calls out Google AI for ‘false’ history

In June 2024, a UNESCO report expressed warnings over the likes of Google Bard (predecessor of Gemini) and OpenAI’s ChatGPT creating ‘false’ content about World War II events and the Holocaust, thanks to the hallucinative nature of AI-generated content. The UNESCO report points out that AI chatbots, like ChatGPT and Google Bard, have been shown to generate harmful misinformation about the Holocaust. These AI models have produced fabricated accounts, such as the non-existent ‘Holocaust by drowning’ campaigns, and false witness testimonies.

Also read: AI hallucination in LLM and beyond: Will it ever be fixed?

UNESCO strongly urged the emphasis on ethical use of AI technologies to ensure the memories of World War II and Holocaust are preserved in their true form for younger and future generations.

3) Google Bard and Gemini mess up during live demo

During its very first demonstration in February 2023, the then Google Bard fumbled badly by providing incorrect information about the James Webb Space Telescope’s discoveries, leading to a $100 billion drop in Alphabet’s market value. A super costly mistake for what was supposed to be Google’s very first public response to ChatGPT at the time and it messed up extremely badly, seemingly to never recover from the mishaps that followed – which show no signs of stopping. 

More recently this year, at the Google I/O conference in May 2024, Gemini’s video search feature again made unforgivable factual errors during a live demonstration, according to a report by The Verge, raising concerns about the reliability of Google’s AI chatbot.

Apart from these high-profile AI controversies, Google also found itself in the middle of a major viewer backlash against its Gemini ads for Paris 2024 Olympics, where the Gemini AI chatbot sparked outrage in an ad that showed a father using Gemini to help his daughter write a fan letter to Olympian Sydney McLaughlin-Levrone. While Google defended the ad, the backlash against it highlighted concerns about the potential negative impacts of AI on human creativity – especially in young children. Then there was also that instance in May 2024 when Google’s AI-generated summaries in search results, known as AI Overviews, contained errors like suggesting users add glue to pizza recipes or eat rocks, prompting Google to scale back the feature.

Well, there’s nothing much to say here apart from the fact that it’s high time for Google to get its Gemini and other AI-related acts together. As much as I don’t want to think about the scary reality of rogue AI, it’s getting embarrassing at this point for Google to constantly keep fumbling on the AI front.

Also read: AI agents explained: Why OpenAI, Google and Microsoft are building smarter AI agents

In the last two years, the world has seen a lot of breakneck advancement in the Generative AI space, right from text-to-text, text-to-image and text-to-video based Generative AI capabilities. And all of that’s been nothing short of stepping stones for the next big AI breakthrough – AI agents. According to Bloomberg, OpenAI is preparing to launch its first autonomous AI agent, which is codenamed ‘Operator,’ as soon as in January 2025. 

Also read: Meet Prithvi, NASA & IBM’s free AI model for better weather prediction

Apparently, this OpenAI agent – or Operator, as it’s codenamed – is designed to perform complex tasks independently. By understanding user commands through voice or text, this AI agent will seemingly do tasks related to controlling different applications in the computer, send an email, book flights, and no doubt other cool things. Stuff that ChatGPT, Copilot, Google Gemini or any other LLM-based chatbot just can’t do on its own. Knowing fully well that I’m getting way ahead of myself, are you ready for J.A.R.V.I.S., Tony Stark’s intelligent AI assistant from Iron Man, or Samantha from Her, a significantly advanced AI operating system than what we’ve experienced till now?

What are AI agents

Simply put, an AI agent is a slightly more advanced AI program that can perform certain autonomous tasks that aren’t just limited to its own base program. ChatGPT or Gemini can write code for you, if you ask for it, but it can’t go and create a website or an app from that code, where the website is live with a domain name or the app published on the app store. An AI agent will be able to do these things – I’m not saying these exact tasks that I suggested above, but these AI agents will have the ability to not just show what needs to be done but also go ahead and do some of that work.

According to Amazon’s official AWS blog, humans will set broad goals for any given AI-related task, where the AI agent will independently choose the best actions it needs to perform to achieve those goals. Amazon further explains how in a customer service scenario, a future AI agent will automatically try to satisfy a calling customer’s query – by looking up internal information, by asking different questions to the human customer, by taking stock of the situation and responding with a solution that solves the calling customer’s problem. In this scenario, the AI agent handles the customer’s call on its own – without passing the call to a human customer support expert. In fact, whether or not to transfer a call to a human customer support expert is determined automatically by the AI agent.

AI agents will be superior to simple AI chatbots thanks to their advanced reasoning capabilities, suggests IBM’s blog post on AI agents. Unlike traditional AI chatbots like ChatGPT or Gemini, which give highly scripted responses to user queries, AI agents will have the ability to plan, think through and adapt to new information, enabling them to handle much more complex tasks with minimal human intervention or supervision.

Difference between AI agents and AI chatbots

There’s a lot of sophistication baked into AI agents, which AI chatbots simply don’t have. One way to look at AI chatbots is that they’re knowledgeable in all the theories of various subjects, whereas AI agents not only have the knowledge but also the expertise to apply their learnings in different applications. Given below are three key differences between AI agents and chatbots…

Being autonomous is the name of the game here. AI agents are self-directed, capable of making their own decisions based on given human instructions by carrying out tasks like scheduling meetings or managing emails – without the need for constant human intervention. This is in stark contrast to AI chatbots like ChatGPT or Gemini that rely on constant user prompts to generate responses, where they lack the ability to initiate actions on their own independently.

Their ability to break down complex tasks and execute is another key differentiator. AI agents are equipped to tackle complex, multi-faceted tasks by drawing on information from diverse sources and making informed decisions. On the other hand, AI chatbots are generally restricted to only providing information or answering queries based on their pre-existing trained knowledge base.

According to experts, another key point of difference is the following: AI agents have the ability to learn from experiences and adapt their behaviour over time to match a set of assigned tasks, enhancing their performance in ever-changing conditions. However, AI chatbots typically lack this level of adaptability, unable to learn anything new apart from what’s there in their existing knowledge base. These are some of the top differences between AI agents and chatbots as we know it.

Different type of AI agents

Just like different AI chatbots have varying levels of competency across different tasks, so do AI agents come in all sizes and shapes – in a matter of saying, of course. An AI agent can be as simple or complex depending on its programming and the quantum of tasks it’s expected to execute. Given this scope, here’s how AI agents are being classified into three main types.

Also read: SLM vs LLM: Why smaller Gen AI models are better

Firstly, there are so-called goal-based agents which are designed to achieve specific objectives by evaluating various action sequences and selecting the most effective path to reach their goals. Unlike simple agents, goal-based agents carefully consider future outcomes and plan their actions accordingly. An example of this goal-based AI agent is a navigation system that identifies the fastest route to a destination by analysing multiple pathways and selecting the one that minimises travel time.

After goal-based agents come what’s known as utility-based agents, which extend the functionality of goal-based agents by not only aiming to achieve a goal but also optimising the quality of its final intended outcome. These AI agents use a utility function to assign a value to each potential outcome, thereby choosing actions that maximise overall satisfaction or performance of any given task. This approach is especially useful when multiple paths can lead to the same goal, allowing the AI agent to select the most advantageous one based on predefined criteria. Imagine a travel booking system that recommends flights not only based on reaching the destination but also considering factors like ticket price, travel time, and layovers to provide the most cost-effective and convenient option – this is what a utility-based AI agent will be able to perform as part of its tasks.

Finally, there are learning agents. These AI agents possess the ability to improve their performance over time by learning from experiences. By continuously interacting with their environment and incorporating feedback, learning agents adapt to new situations and refine their decision-making processes, making them suitable for dynamic and complex domains. There are also something known as hierarchical agents, which are nothing but a group of AI agents arranged in multiple tiers. In such a hierarchical structure, higher-level agents break down complex tasks and assign them to individual lower-level AI agents. These lower-evel agents run their tasks independently and hand over their results to the higher-level agents up the value chain.

Who all are developing AI agents

All the big movers and shakers of the AI industry are planning to release their version of AI agents for the public very soon in 2025, if they haven’t done it already by late 2024.

As I mentioned earlier, OpenAI’s reportedly working on getting their AI agent, codenamed Operator, out into the open for everyone to check out by January 2025. The AI agent is expected to be capable of autonomously operating certain tasks within your computer, like booking flight tickets and implementing code, among other things. Google seems to be working on several AI agent projects, one of which is known as Project Jarvis, which recently leaked on the Chrome Web Store. This AI agent will supposedly reside within Google Chrome browser, with the ability to not only automate and execute tasks within the browser but also operate other apps on the host PC or computer. While there’s no set release date for Project Jarvis yet, however, Google’s Gemini 2.0 AI model is expected to have AI agents built-in to offer enhanced capabilities – it’s expected to release later this year in 2024.

Microsoft is also working aggressively on AI agents, something that it had announced earlier in the year in 2024. According to its official blog, new capabilities in Copilot Studio will allow Microsoft customers to create powerful autonomous AI agents. Some of these demonstrations are in public preview at the moment, where AI agents can draw upon work or business data from different Microsoft Office 365 apps to undertake a variety of assistive tasks – like IT help desk, employee onboarding, coordinating sales and service, and more.

Anthropic, an AI startup competing with OpenAI, has already released its AI agent for people to try. According to Techcrunch, Anthropic has made significant upgrades to its Claude 3.5 Sonnet AI model which now lets it use the host computer – yes, it can interact with computers in a way that mimics humans. It can move the cursor around the screen, click on apps and buttons, and it can potentially interact with other softwares and programs installed in your PC to autonomously execute various tasks. How scary and cool is that?!

If the world wasn’t prepared for Generative AI back in 2022, then let me tell you it’s certainly not prepared for AI agents and all the various ways it can impact our lives – for better or worse. Let’s hope these AI agents don’t turn out to be the dystopian versions as depicted in an iconic movie 25 years ago, for your sake and mine, eh?

Also read: Meta AI manifesto: The AI-assisted resurrection of Mark Zuckerberg

After historic setbacks in antitrust cases related to its online search and advertising empire in recent months, Google’s now facing an unprecedented legal challenge to its Google Play Store policy – and why this matters for India as well.

Google ordered to do three key things

Earlier on October 7, US District Judge James Donato issued an injunction set to take effect on November 1, mandating significant changes to Google’s Play Store policies. The order stems from a legal battle with Epic Games, the creator of the popular game “Fortnite,” which successfully argued that Google was monopolising app distribution and payment methods on Android devices, violating antitrust laws in the US.

Also read: Google’s AI Summaries: Where everyone loses out. Eventually.

The injunction issued by the US District Judge requires Google to do the following things: Google should officially allow users to download apps from third-party platforms or app stores on Android devices, a practice known as “sideloading.” The US court injunction also prohibits Google from blocking the use of competing in-app payment systems. 

Additionally, Google can no longer pay device manufacturers to preinstall its Play Store nor share revenue generated from the store with other app distributors. These changes are solely aimed at reducing Google’s excessive and overarching control over the Android apps and devices ecosystem, which the US court found to be unfairly limiting competition.

In response to the US court’s instruction, Google has argued that such injunctions could harm its business and raise security, privacy, and safety concerns within the Android ecosystem, according to a Reuters report. As of late October, Google has temporarily avoided major changes to its Play Store as a US federal judge blocked the original injunction in its ongoing legal battle with Epic Games, except one key point: Starting November 1, Google has been barred from tying payment or revenue-sharing agreements to exclusive Play Store pre-installation on devices. This restriction will be in effect for three years, ending November 1, 2027.

Google’s anti-competitive behaviour in India

In India, similar concerns have been raised about Google’s dominance in the app distribution market. The Competition Commission of India (CCI) has previously fined Google for abusing its dominant position in the Android mobile device ecosystem. In October 2022, the CCI imposed a penalty of ₹1,337.76 crore on Google for anti-competitive practices related to Android mobile devices. 

In March 2024, the CCI ordered an investigation into Google’s Play Store billing policies after Indian internet companies complained about the high commission structure, ranging from 11-30% on in-app purchases made by their users. For context, credit card payment fees generally range from 1-2%, while UPI transactions are free, making Google’s rates disproportionately higher, as per an Economic Times report. To put this into further perspective, affiliate commissions – which are typically paid for referring customers – range between 2-10%, depending on the category of purchase. This raises the question: why should App Store fees be so high when affiliate commissions are much lower? The companies which filed cases include Shaadi.com, BharatMatrimony, Kuku FM’s parent company Mebigo Labs, the Indian Broadcasting and Digital Foundation and the Indian Digital Media Industry Foundation, among others. 

Also read: Google’s antitrust cases in India: A brief history

Subscription-based apps also face a double burden – Google charges approximately 30% on the first year of subscriptions and around 15% on renewals after the first year. Many app developers argue this is unfair, as companies should only be charged for discovery and distribution, not for future renewals. The industry sentiment is that a fair rate would be closer to 5%, particularly for long-term subscribers. Moreover, even if a customer uses UPI – where transactions are free for merchants – to pay for in-app purchases, app publishers are still charged these high commissions by Google, further amplifying the unfairness of the current system. In a 21-page order, the Indian antitrust regulator said it believes that Google may have violated the Competition Act and has decided to launch a detailed investigation into the matter. Although the investigation is ongoing, Indian companies have yet to receive compensation or any tangible relief.

Currently, there are few notable alternatives to the Google Play Store in India. However, options like GetApps and Galaxy Store exist but lack the scale and traction of Google Play. If the US ruling sets a precedent, India could witness more app stores entering the competition, offering developers a fairer marketplace.

In simple terms, Google operates the Play Store, the primary platform from which Android users can download apps. While Android is an open-source operating system, Google’s Play Store has become the dominant app distribution channel. Developers who wish to reach a wide audience often feel compelled to distribute their apps through the Play Store and use Google’s in-app payment system, which charges a commission on transactions. This has raised concerns about fairness, as developers argue they should not be forced to use Google’s in-app payment system for every transaction. Instead, they believe Google should only charge a referral fee for discovery and download, allowing developers to use their own payment processors for in-app purchases, reducing the financial burden on app creators.

Many developers and media companies are frustrated with additional restrictions such as Google’s prohibition on certain types of links and ads, including retail links for news apps. For instance, news apps are not allowed to include retail links without facing penalties or removal from the Play Store, limiting their ability to generate revenue through affiliate marketing or sponsored content. Similarly, media companies are barred from running certain types of ads that compete with Google’s ad services, further restricting their monetization options. These prohibitions have sparked criticism, with industry stakeholders arguing that such practices stifle competition and unfairly favour Google’s own services. This setup has been criticised for limiting competition and imposing unfair terms on app developers.

According to a June 2024 report, the Indian Ministry of Information and Broadcasting (MIB) held a meeting to discuss the imbalance in bargaining power between Big Tech companies and digital news publishers, with the latter being heavily reliant on the former for traffic and advertising revenue. The Digital News Publishers Association (DNPA) is believed to have advocated for a revenue-sharing mechanism with Big Tech companies, similar to Australia’s News Media Bargaining Code, France’s settlement with Google and Canada’s Online News Act. The meeting wasn’t conclusive on whether a legal framework is required to address the issues raised by news publishers or if a case-by-case redressal approach is better instead.

Impact on publishers and journalism

However, a new pressing concern is that AI-generated content and AI-driven search overviews are further killing publisher traffic by displaying answers directly to users, bypassing news sites altogether. This is seen as unethical, as AI systems are using publisher content without permission or compensation, raising large legal implications for copyright and content ownership. This practice not only undermines publishers’ revenue but also violates intellectual property rights, sparking calls for urgent legal measures to protect the rights of content creators in the digital age.

Additionally, Google’s increasing use of “zero-click searches” is compounding the issue. A zero-click search occurs when a user’s query is answered directly on the search results page, often through snippets, without the need to click through to the original publisher’s site. 

According to Search Engine Land, a recent study by Semrush has revealed that nearly 60% of Google searches in 2024 resulted in no or zero-clicks. In the US and EU, over 50% of searches end without a user clicking on a result. While Google’s own properties receive a substantial portion of clicks, only 36% of clicks go to external websites. This suggests that a significant number of users are finding the information they need without leaving Google’s platform.

This practice is turning Google from a search engine into a content provider, significantly reducing the traffic that would otherwise go to publishers’ websites. The combination of AI overviews and zero-click searches is raising concerns about the future sustainability of digital media businesses.

By taking traffic away from publishers and disincentivizing them, there is a risk that publishers will start to look for alternative traffic sources, which may not be as reliable or lucrative as Google’s platform. This could lead to a reduction in resources for quality journalism, as publishers would have less incentive to invest in high-quality content if their revenue models continue to erode. Ironically, this decline in journalism quality would eventually affect Google itself, as lower-quality content would mean poorer search results, undermining the core value of Google’s search engine.

Indian legal experts weigh-in

Reacting to the ruling against Google by a US District Judge, cyber law experts believe similar ruling here would boost small developers and startup ecosystem in India, a digital economy dominated by US-based big tech firms. 

“A ruling like this would foster an ecosystem where small developers and startups can compete fairly without being unduly burdened by the policies of dominant platforms. This is especially crucial for India’s startup ecosystem, where companies rely heavily on platforms like Google Play and Apple’s App Store to reach users,” according to Adv (Dr) Prashant Mali, a Mumbai-based practising lawyer and AI & Cyber Public Policy thought leader. “The Google vs. Epic Games ruling raises questions about whether consumers are being deprived of choice in digital marketplaces,” Adv (Dr) Prashant Mali further adds.

“The recent US court verdict is widely welcomed, as it opens the doors for fair and free market practices,” Adv Ashraf Ahmed Shaikh, Mumbai High Court, reacted while welcoming the injunction against Google, further explaining how “this verdict will prevent exploitation due to single monopoly.”

According to Adv (Dr) Prashant Mali, “India should take measures to ensure that users have access to a wider range of payment systems and platforms. This would encourage innovation, allowing smaller players to offer competitive services.”

It’s all about keeping public interest above private interest, or else Big Tech companies may become dominant and would interfere with the basic structure of liberty and freedom, warns Adv Ashraf Ahmed Shaikh. “This may give rise to authoritarianism and exploitation, therefore checks and balances are needed. No one should get a monopoly or else they start dictating terms. As we say absolute power corrupts absolutely,” says Adv Ashraf Ahmed Shaikh.

“With India’s digital market on the rise, regulations ensuring fair play for all developers can help local companies thrive. This can boost India’s goal of becoming a global tech hub by enabling homegrown businesses to grow without fear of being stifled by monopolistic practices,” says Adv (Dr) Prashant Mali. 

Highlighting India’s Digital Personal Data Protection Act (DPDP Act) of 2023 and Digital Competition Bill (DCB) of 2024, which aims to regulate digital enterprises and their practices to promote fair competition and protect Indian consumers, “India has already taken crucial steps in this direction,” sums up Adv Ashraf Ahmed Shaikh. India has taken crucial steps, but faster legislative and enforcement actions are needed to catch up with the rapidly evolving tech sector. Ultimately, though, is this enough to keep the power of Big Tech companies in check?

As the global conversation on antitrust regulations intensifies, Indian authorities have an opportunity to reassess and strengthen their policies to prevent abuse of market dominance and promote fair trade practices in the technology and media sector.

Also read: Google for India 2024: All eyes on AI