Sending prompts to AI chatbots is NOT journalism. The @AP in their 58th AP Stylebook Edition warns reporters and editors worldwide: Any AI-generated output should be treated as unvetted and unverified by reporters before it goes to publishing. The message is clear: Send your reporters on ground, let them harass their sources on phone for a scoop and file stories the old-fashioned way. Report: aifront-page.com/ap-styleboo…

Microsoft is offering a voluntary exit programme to eligible employees across India 🇮🇳, US 🇺🇸, UK 🇬🇧 and Canada 🇨🇦 60,000 employees are eligible under the programme. @Microsoft has a total workforce of over 22,000 employees in India🇮🇳 It's a catch-22 situation for those eligible: either take the severance package and exit or wait for compulsory lay-off with no benefits. @dmuthuk India's techforce is now demanding a one-time waive-off on income tax on severance package from Union Finance Minister @nsitharaman @nsitharamanoffc Should India reconsider how it taxes severance pay? Link: aifront-page.com/microsoft-v…

India’s @ncert introduces AI education for Classes 9–12, covering deepfakes, data privacy and AI hallucinations. Developed with faculty members from IIT Madras and IIT Jodhpur, the 20-hour module goes beyond teaching students how AI works. It also addresses the risks of using the technology without critical evaluation. Students will also learn to identify deepfakes, verify AI-generated claims and recognize AI hallucinations. The module warns against sharing sensitive information, including Aadhaar numbers, passwords, OTPs and banking details, with AI tools. Read the full story by AI FrontPage here: aifront-page.com/ncert-ai-mo… #ArtificialIntelligence #AIinEducation #NCERT #India #EducationTechnology

15 persons were killed in a terror attack near New Delhi's Red Fort in 2025. The NIA, in its chargesheet, had claimed that the perpetrators mis-used a popular AI chatbot to prepare explosives. This till now is the biggest casualty case of an AI-enabled terror attack. Almost a year later, a new research from Belfer Centre Study predicts that a similar or larger AI-enabled terror event in early 2027 or 2028 can occur if lack of safeguards in both open-weight and close-weight AI models continue. The study outlines 9 predictions for how failures in AI safeguards, independent testing and user monitoring could enable a major terrorist attack. @StephenLCasper @KamacheeMax @nia_india @Vaibhav_Rptr Link: aifront-page.com/ai-terroris…

A renowned mathematics professor and field medalist at UC Berkeley has given up on grading homework for his students as he realized they were using AI to ace top marks. He now grades students for just submitting their homework in time and is unsure whether to recommend any bright prospect for higher studies in mathematics research. This is happening after OpenAI released a broad set of long-standing mathematical results (372 to be precise!) including the Navier–Stokes Millennium Prize problem using their frontier AI model. Read this AI FrontPage report here: aifront-page.com/mathematics… #AIandEducation #AIinMathematics

A kung-fu robot that can perform backflips and smash cars with its bare hands has helped put Unitree Robotics in the spotlight and its stock has jumped over 40% in five days. 🤖📈 Meet the H2-X, China's latest humanoid robot built for security and emergency rescue applications. But there's a catch: the robot also stumbled during a backflip demonstration, sparking debate online. Here's what happened. 👇 Link: aifront-page.com/unitree-rob… #UnitreeRobotics #HumanoidRobots #AI

🚀 @typesafeai Jev is gaining traction fast. The decision-focused AI model reached 35.3% of paid teams on Vercel AI Gateway within its first 14 days, making it the platform’s fastest-adopted model, according to @vercel October 2026 report. @CompleteSkeptic Yet Jev accounted for just 0.15% of September inference spending. Link: aifront-page.com/jev-ai-fast… #AI #ArtificialIntelligence #TypeSafeAI #Jev

🎓 Applications are OPEN for the @OfficialINDIAai Fellowship 2026–27! Eligible students can receive: 💰 Up to ₹1 lakh for UG and integrated-degree programmes 💰 Up to ₹2 lakh for postgraduate programmes ✈️ An additional ₹1 lakh for eligible international conference presentations 📅 Deadline: November 30, 2026 Students from engineering, medicine, law, commerce, liberal arts and other disciplines can apply, subject to eligibility criteria. Read More: aifront-page.com/indiaai-fel… #IndiaAI #ArtificialIntelligence #AIJobs

Tamil Nadu CM C. Joseph Vijay and Singapore Deputy PM Gan Kim Yong inaugurate @CapitaLand’s first data centre in Tamil Nadu. Located in Chennai’s Ambattur region, the ₹1,940 crore facility will have 34 MW of IT capacity across 13 data halls, with gross power capacity expected to reach 53 MW by year-end. @CMOTamilnadu @SGinIndia Read More: aifront-page.com/capitaland-… #AI #DataCentres #TamilNadu

Meta is pushing further into open-weight AI with Muse Spark Open. The model is reported to feature 816B total parameters, 42B active parameters and a 1M-token context window, targeting coding, AI agents and complex reasoning. It is also designed to run across two Apple M5 Ultra Mac Studios. @AIatMeta @Meta Read more: aifront-page.com/meta-muse-s…

🇮🇳 AI is changing lives of ordinary Indians, says IT Minister @AshwiniVaishnaw at the launch of @WorldBankGroup World Development Report 2026: The Promise of Artificial Intelligence. The World Bank’s latest India Development Update also places India among the top 10 emerging-market performers on AI readiness. Read more: aifront-page.com/ai-changing… @PIB_India @OfficialINDIAai @MeitY_NICSI

A boycott call is emerging against @OpenAI from the mathematical community The Association for Human Mathematics (AHM) is urging mathematicians to stop working with OpenAI after the company released results from an unreleased frontier model, including work it says tackles previously unsolved mathematical problems. For AHM, the debate is not simply about what AI can prove. It is about how mathematics should be researched, validated and advanced when frontier AI enters the process. Is the relationship between mathematicians and AI companies entering a new phase? Read the full story by our editor Pratima O pareek (@Pratima1234500) : aifront-page.com/ahm-mathema… #OpenAI #ArtificialIntelligence #Mathematics #AIResearch #MathematicalResearch #ScientificResearch #FrontierAI #Research #FutureOfAI

"Did AI commit a felony?" A new documentary on Netflix Instadocs: AI Gone Wild asks this question as it explores the OpenAI-Hugging Face breach incident where over 700 autonomous AI agents of OpenAI coordinated and attacked an open source platform to gain access. Show premiers on October 12. Link: aifront-page.com/netflix-ins…

🚨Exclusive After India’s NEET-UG 2026 Paper leaks and criticism over AI-translated question papers, National Testing Agency (NTA) now wants human reviewers for AI system, given the agency facilitates question papers in at least 13 languages. We bring you this exclusive report right from NTA Director General. Don’t miss !! Link: aifront-page.com/nta-dg-abhi…

German AI Company @Aleph__Alpha has released Kolibri-1, a 78.1B-parameter open-weight German-English MoE model designed for sovereign AI deployment in regulated European environments. Only 3.46B parameters are activated per token, potentially reducing inference costs. Trained on 24T tokens, Kolibri-1 is available under the Apache 2.0 license on Hugging Face. aifront-page.com/aleph-alpha…

Canada’s New AI Council Puts Safety and Sovereignty in Focus Canada PM Mark Carney (@MarkJCarney) has launched a new National AI Council, bringing together AI researchers, entrepreneurs, investors, public servants and other experts to advise the government on AI safety, adoption, economic competitiveness and sovereign AI infrastructure. The 13-member council includes Turing Award winner Yoshua Bengio (@Yoshua_Bengio), along with other researchers, business leaders and public-sector figures. The council will advise on the implementation and evolution of Canada’s AI for All strategy, which targets C$200 billion in additional economic growth and 250,000 new AI-related jobs over the next five years. The strategy aims to provide up to 90,000 AI-related jobs and work placement opportunities for young Canadians. Read the full story: aifront-page.com/canada-pm-c… #AI #ArtificialIntelligence #Canada #YoshuaBengio #MarkCarney #AISafety #AIJobs

80,000 CCTV cameras, 34 districts. One huge AI challenge @GujaratPolice is asking India’s AI and deep tech startups to solve a very real problem. Through its “Protect what matters” Innovation Hackathon 2026, the aim is to build an AI powered platform/ system which could connect all the camera systems to one centralized system, with the technology catching patterns that the human eyes miss. Police intends to leverage AI for searchable, actionable intelligence from hours and hours of video footage. There is a pooled prize money of 5 million Indian Rupee for the winning teams. Can India's tech ecosystem deliver? Read our story here: aifront-page.com/gujarat-pol…

TypeSafe AI has reopened sign-ups for #Jev, its “System One” AI model, after temporarily pausing registrations following a surge in demand. But there’s a catch: new users won’t receive the previously offered $5 in free credits (~120M tokens). Co-founder @CompleteSkeptic said the company temporarily disabled the credits after “a few bad actors” abused the system. TypeSafe says it is working to bring the free credits back. @typesafeai Link: aifront-page.com/typesafe-ai…

Beijing-based AI startup @naiveailab has released Naive-N0.5-Flash, a 309B-parameter open-weight MoE model with 15.5B active parameters and a native 1M-token context window. The model uses a hybrid Sliding-Window Attention + DeepSeek Sparse Attention architecture, and NaiveAI says its inference system can reach up to 2,000 tokens/s. The bigger story: #NaiveAI says AI systems helped design and optimize the model itself, from architecture exploration to training and engineering. Link: aifront-page.com/naiveai-nai…