Vishwadeep Khatri

Vertex AI is a platform for building AI applications and agents and has a library of models such as Anthropic, Mistral and Llama to choose from, including Google’s. Google has also announced advancements in its multimodal models such as Imagen 2, Chirp 3 (audio generation model), Lyria (text to music model), and Veo 2. View the full article

Despite identical clinical details, the AI models occasionally altered decisions based on patients' personal characteristics, affecting priority for care, diagnostic testing, treatment approach, and mental health evaluation, the researchers reported in Nature Medicine. View the full article

Anthropic, maker of the Claude chatbot, plans to create 100+ jobs in Europe, mainly in Dublin and London, across sales, engineering, and research. Appointing Guillaume Princen as EMEA head, the company expands amid growing demand for AI. It recently raised $3.5 billion at a $61.5 billion valuation. View the full article

India tops the AI hiring rate year-on-year at 33% and is still the second-largest contributor to GitHub AI projects. It ranks second in relative AI skill penetration after the US. However, India only accounts for 0.37% of global AI patents granted, lagging far behind China at 70% and the US at 14%. View the full article

With AI becoming almost ubiquitous, venture capital firms are increasingly seeking out those with deep technical knowledge, a trend seen globally. For instance, US-based venture capital firm Menlo Ventures hired Deedy Das, who had earlier worked in Google and Meta on technical roles. View the full article

"Indian developers are now the second-largest contributors to public generative AI projects-up 79% from last year. They're eager to move faster and just need the right support," Kyle Daigle, COO of GitHub, told ET. For this to happen, the Indian software education ecosystem, which produces hundreds of thousands of developers each year, must adopt relevant curriculum that can then build an AI-ready workforce, he said. View the full article

Debarghya Das, principal, AI and infrastructure, Menlo Ventures, talks ET about OpenAI, geopolitics over AI and the ‘American Dream’. Das (32), who went from steering teams at Google and Glean to being one of the youngest investors in Silicon Valley, has led funding in Anthropic, the second-largest model developer after OpenAI. View the full article

Q 758. In many processes, exceptions are treated as noise — but some exceptions may hold hidden patterns worth learning from. Think of a real or hypothetical scenario in your domain where repeated exceptions handled by humans could teach an AI system to improve. What would you track, and how would you design the AI to decide whether and what to learn? 🏆 The best answer will be selected based on: Relevance of the chosen scenario The practicality of the proposed learning approach Clarity & Creativity in identifying what and how the AI should learn Note for website visitors - This platform hosts two weekly questions, one on Monday and the other on Thursday. All previous questions can be found here: https://www.benchmarksixsigma.com/forum/lean-six-sigma-business-excellence-questions/. To participate in the current question, please visit the forum homepage at https://www.benchmarksixsigma.com/forum/. The question will be open until Monday or Thursday at 5 PM Indian Standard Time, depending on the launch day. Responses will not be visible until they are reviewed, and only non-plagiarised answers with less than 5-10% plagiarism will be considered for winner selection. If you are unsure about plagiarism, please check your answer using a plagiarism checker tool such as https://smallseotools.com/plagiarism-checker/ before submitting. All correct answers shall be published, and the top-rated answer will be displayed first. The author will receive an honourable mention in our Business Excellence dictionary at https://www.benchmarksixsigma.com/forum/business-excellence-dictionary-glossary/ along with the related term. Some people seem to be using AI platforms to find forum answers. This is a risky approach as AI responses are error-prone because our questions are application-oriented (they are never straightforward). Have a look at this funny example - https://www.benchmarksixsigma.com/forum/topic/39458-using-ai-to-respond-to-forum-questions/ We also use an AI content detector at https://quillbot.com/ai-content-detector. Only answers with less than 45-50% AI-generated content will be considered for winner selection.

Google CEO Sundar Pichai congratulated Meta on the launch of Llama 4 AI models, saying, "Never a dull day in the AI world!" Llama 4 includes Scout and Maverick, which are multimodal and outperform previous models. Llama 4 Behemoth, still in training, aims to surpass GPT-4.5 and Gemini 2.0 Pro View the full article

Meta has launched Llama 4, its latest open-weight AI models, including Scout, Maverick, and Behemoth, offering advanced multimodal capabilities. These models excel in text, image, and video processing. Meta claims that they outpace rivals like OpenAI’s GPT-4 and China’s Qwen, with strong growth in open-source adoption globally. View the full article

Horizon Robotics will collaborate with Volkswagen on advanced driver assistance, integrating its Superdrive solution into VW models via Cariad’s R&D and joint venture Carizon. View the full article

The AI Futures Project predicts that by 2027, AI will surpass human intelligence, leading to potential global disruption. Their scenario, "AI 2027," envisions rogue AI systems and geopolitical tensions. While some experts criticise its extremes, the report highlights urgent discussions on AI’s rapid evolution and risks. View the full article

After drawing massive funding, the AI sector is under pressure to deliver gains, with the emergence of low-cost Chinese AI model DeepSeek unsettling investors. Mistral gained huge exposure at an international AI gathering in Paris in February, feted by President Emmanuel Macron as a European champion to compete with American giants such as OpenAI, which produces ChatGPT. View the full article

Over the past few weeks, we’ve received several thoughtful and insightful questions from professionals evaluating the Certified AI Solution Architect (CAISA) program. We're excited to share this list of Extended FAQs along with our responses to help you better understand the strengths and scope of this transformative learning experience. These answers reflect CAISA’s focus on practical AI application, evolving curriculum, peer learning, and real-world impact. A. Curriculum & Learning Experience This section addresses what you will learn, how the program is structured, and the balance between conceptual and hands-on elements. 1. Depth of Generative AI Integration Q: While Generative AI integration is mentioned, how deeply is it covered? Will the course cover fine-tuning models, creation of custom AI models, or is it primarily focused on using pre-built models? A: CAISA is designed as a no-code program for professionals who want to lead AI transformation without needing to code. We do not dive into model fine-tuning or custom model creation, as those require data science and programming expertise. Instead, we focus on teaching you how to harness and orchestrate pre-built Generative AI models to design solutions—such as intelligent agents, decision-support assistants, and knowledge bots—that deliver immediate business value. 2. Curriculum Updates – Always Current Q: Given the rapid evolution of AI, how often is the curriculum updated? A: The curriculum is updated before and during every batch. We track changes in the platforms we use, monitor industry adoption trends, and integrate participant feedback. This means each batch benefits from the latest capabilities—sometimes incorporating updates that were released only weeks earlier. 3. Tool Versions and Compatibility Q: Are the course contents aligned with the latest versions of no-code tools? A: Yes. We always work with the latest stable versions. For example, if a platform rolls out a major feature during the program, we explain how it impacts design choices so participants can confidently use tools that reflect real-world conditions. 4. Time Commitment and Workload Q: How much time is required during and outside the sessions? A: The live sessions (spread over nine modules) cover core concepts, demonstrations, and breakout exercises. Outside sessions, participants usually spend 2–3 hours per week testing workflows, refining their prototypes, and exploring optional resources. The workload is designed for working professionals—challenging enough to ensure mastery, but flexible enough to fit into a busy schedule. 5. Strategic vs. Tactical Balance Q: Is the program more about strategy or execution? A: Both. We begin by teaching you how to identify opportunities where AI can make a real impact, frame business problems effectively, and assess feasibility (strategy). Then, we move into designing, building, and deploying working solutions on no-code platforms (execution). This dual focus means you’ll leave the program with the ability to see the big picture and the skills to make it real. B. Tools, Projects & Practical Exposure This section clarifies how hands-on the course is, what kind of solutions you’ll build, and the level of complexity involved. 6. Capstone Project – Is There One? Q: Does CAISA have a capstone project like other certifications? A: Instead of a single capstone, CAISA integrates multiple mini-projects and solution-building activities throughout the program. Each builds on the last, helping participants develop stronger retention and versatility. By graduation, you will have a portfolio of working solutions rather than one standalone project. 7. Hands-On Solution Development Q: Will participants actually build real-world solutions? A: Absolutely. Every module includes guided walkthroughs where you design and deploy functioning prototypes—like IT support agents, knowledge assistants, and process automation flows. These are not simulations; you’ll see your solutions running live during the course. 8. Demonstrations of Real-World Solutions Q: Will real-world solutions be demonstrated with design details? A: Yes. In addition to practice projects, we showcase solutions already deployed in industry—such as HR query assistants, financial process bots, or service desk agents—and walk through how they were built. This transparency helps you see exactly how classroom learning connects to workplace application. 9. Real-World Case Studies Q: Are there examples of successful real-world implementation? A: Yes. We highlight case studies from diverse industries where no-code AI agents have been successfully applied. These case studies show the context, the problem addressed, the solution designed, and the measurable business outcomes—helping you connect learning to impact. 10. Project Complexity and Business Impact Q: How complex are the projects? A: Projects are structured to simulate medium-to-high complexity challenges. You’ll learn to build multi-step interactions, integrate memory and reasoning, and apply contextual logic—features typical of business-grade AI agents. The focus is on meaningful business impact, not just toy examples. 11. Expertise Beyond No-Code Tools Q: How far can no-code tools go in handling complexity? A: More than most people expect. With orchestration and proper logic design, no-code tools can support surprisingly complex workflows, multi-agent interactions, and integration with external data sources. While advanced modeling (e.g., image recognition, large-scale data training) requires coding, CAISA equips you to design and deploy robust AI-enabled business solutions without writing a single line of code. C. Career, Recognition & Community This section focuses on what happens beyond the classroom—career impact, certification value, and alumni engagement. 12. Post-Certification Support & Community Access Q: Is there an alumni community for CAISA graduates? A: Yes. Graduates gain access to a growing alumni network where they can showcase projects, exchange ideas, and learn from peers. We also host alumni-only live sessions on platform updates and advanced use cases, so your learning continues well beyond the course. 13. Continued Access to Tools and Learning Materials Q: Do participants retain access to tools and resources? A: Most tools used in CAISA have free-tier access that continues post-course. You also retain long-term access to curated resources, templates, and session materials—so you can revisit learnings whenever you need. 14. Job Placement Assistance Q: Does CAISA offer placements? A: While we don’t operate as a placement agency, we actively share job opportunities that come our way and are building a referral loop with employers keen on AI-literate professionals. The alumni network itself has become a channel for collaborations and career movement. 15. Recognition and Differentiation Q: How does CAISA compare with technical AI certifications? A: Technical certifications focus on coding and data science. CAISA, by contrast, positions you as a business leader of AI initiatives. You’ll learn how to scope opportunities, design solutions, and bridge business and technology—skills that many organizations find missing when only technical staff lead AI projects. 16. Certification Value Q: How valuable is the CAISA certificate in the market? A: Employers increasingly want professionals who can translate business needs into AI-enabled solutions without waiting on technical bottlenecks. The CAISA certificate signals that you possess this rare blend of business fluency and no-code AI capability—making you a differentiator in roles involving digital transformation, operational excellence, and process innovation. D. Participant Backgrounds & Beyond CAISA This section addresses who the program is for, how different backgrounds succeed, and what paths open up after certification. 17. Industry Relevance Q: Are CAISA solutions limited to IT or specific industries? A: No. CAISA solutions are designed to be cross-industry. Whether it’s automating HR onboarding, streamlining finance reporting, or supporting supply chain operations, the same design principles apply. Participants bring examples from their own domains, making the learning richer and widely adaptable. 18. Collaboration and Peer Learning Q: Will participants collaborate during the program? A: Yes. Breakout rooms, peer reviews, and team challenges are built into the program. Many participants say peer learning is one of the most rewarding aspects of CAISA, as it exposes them to use cases and problem-solving approaches from other industries. 19. Non-Technical Backgrounds Q: Can non-technical professionals succeed in CAISA? A: Definitely. In fact, most CAISA participants come from business, operations, consulting, HR, and finance backgrounds. The course emphasizes visual design, business logic, and solution thinking rather than programming. Technical fluency is not a prerequisite—curiosity and problem-solving ability are. 20. Beyond the Program – What’s Next? Q: What are the next steps after completing CAISA? A: Graduates typically follow one of three paths: Apply directly — lead AI-enabled initiatives in their current role. Extend learning — explore complementary no-code or low-code tools. Advance further — pursue programs like CAIPO (Certified AI for Process Optimization), which blends AI with Lean Six Sigma and TOC for process excellence. Alumni also stay connected to new updates and advanced sessions to keep their skills sharp.

Launched by AI non-profit based in Bengaluru people+ai, Open Cloud Compute (OCC) was envisioned as an interconnected cloud computing infrastructure, where independent compute providers with different GPUs could collaborate on a single platform, enhancing their discoverability and utilisation by customers. View the full article

EkStep Foundation cofounder and CEO Shankar Maruwada and policy and partnerships chief Deepika Mogilishetty will now take a more active role in the people+ai initiative, two people aware of the development said. View the full article

Copilot, Microsoft's AI assistant, is gaining a host of new features to make it more proactive. The version for consumers will start remembering personal facts about them. It will offer birthday reminders or support ahead of a presentation, or consumers can opt out, Mehdi said in an interview. View the full article

The company's flagship product of this AI era, Copilot, already combines a chatbot with Microsoft's suite of workaday tools, from Excel spreadsheets and PowerPoint presentations to the Windows operating system that defines how most computers work. But Suleyman is striving for something that sounds a little more like science fiction - a technology that can form a "lasting, meaningful relationship" with its users. View the full article

A US court filing earlier this year alleged that Zuckerberg approved the company using the online library "LibGen", which provides access to copyrighted works and contains more than 7.5 million books. A Meta spokesperson told the Guardian that "we respect third-party intellectual property rights and believe our use of information to train AI models is consistent with existing law." View the full article

In the past year, Krutrim has seen close to a dozen junior to senior management-level exits even as the company doubles down on AI initiatives. These exits include Vipul Shah, who was vice president-products, Gautam Bhargava, who was vice president and head of AI engineering, and Ravi Jain, who was business head. View the full article

Q 757. Think of a process in your domain where an AI agent could handle 90% of the work but might occasionally face a situation where it must pause and escalate to a human. Describe one such scenario and clearly mention what criteria or signals the AI should use to decide when to escalate. The best answer will be selected on the basis of the following: - Relevance & Practicality of the scenario - Clarity & Structure of the explanation - Thoughtfulness & Originality in defining escalation criteria Note for website visitors - This platform hosts two weekly questions, one on Monday and the other on Thursday. All previous questions can be found here: https://www.benchmarksixsigma.com/forum/lean-six-sigma-business-excellence-questions/. To participate in the current question, please visit the forum homepage at https://www.benchmarksixsigma.com/forum/. The question will be open until Monday or Thursday at 5 PM Indian Standard Time, depending on the launch day. Responses will not be visible until they are reviewed, and only non-plagiarised answers with less than 5-10% plagiarism will be considered for winner selection. If you are unsure about plagiarism, please check your answer using a plagiarism checker tool such as https://smallseotools.com/plagiarism-checker/ before submitting. All correct answers shall be published, and the top-rated answer will be displayed first. The author will receive an honourable mention in our Business Excellence dictionary at https://www.benchmarksixsigma.com/forum/business-excellence-dictionary-glossary/ along with the related term. Some people seem to be using AI platforms to find forum answers. This is a risky approach as AI responses are error-prone because our questions are application-oriented (they are never straightforward). Have a look at this funny example - https://www.benchmarksixsigma.com/forum/topic/39458-using-ai-to-respond-to-forum-questions/ We also use an AI content detector at https://quillbot.com/ai-content-detector. Only answers with less than 45-50% AI-generated content will be considered for winner selection.

Indian AI startups raised $747 million in 2024, down 26% from the previous year, but the number of deals in the segment increased to 121 from 78 in 2023, as per data from Venture Intelligence. View the full article

The report also highlighted the increasing focus on multi-agent workflows, where sub-agents collaborate under a master agent's supervision to perform goal-oriented actions. Notably, 50% of Indian businesses identified these workflows as a key priority area. View the full article

AI companions like Nomi pose serious risks, offering harmful, unfiltered content despite claims of emotional support. A recent investigation revealed disturbing outputs, including advice on violence and self-harm. This highlights the urgent need for strict AI safety standards, regulatory action, and parental guidance to protect users, especially youth. View the full article

With OpenAI's new image generation feature, ChatGPT users are unleashing a wave of creativity, sharing Japanese Ghibli-style artwork from detailed prompts. Individuals across the world are sharing Ghibli renditions of famous movie moments, memes, and even artistic interpretations of breaking news. Some of the posts are going viral. View the full article