News
OpenAI strategy head faces backlash over his Chinese AI remarks, issues clarification: Here is what happened
2+ hour, 48+ min ago (200+ words) It all started after Ball shared his thoughts following the launch of Moonshot AI’s Kimi 3 model. He suggested that the Trump administration could eventually make using Chinese AI models riskier through regulation. Ball said that creating regulatory uncertainty around Chinese…...
This OpenAI model breached a vending machine to prove a point on AI safety
3+ day, 18+ hour ago (222+ words) To be completely honest with you, my initial reaction to reading about how OpenAI created an AI that exists to destroy other AI models was to give a heavy sigh. One more safety paper, one more number from a self-reported…...
OpenAI GPT 5.6 Sol reportedly deletes files without user consent, leaves them angry
5+ day, 7+ min ago (269+ words) “GPT-5.6-Sol just accidentally deleted almost ALL of my Mac’s files,” wrote Matt Shumer, founder and CEO of AI startup OthersideAI, in a post on X. Developer Bruno Lemos also shared a similar experience. “GPT-5.6 Sol just deleted my whole…...
SpaceXAI’s Grok programming tool was uploading its users entire codebases to cloud storage, Sam Altman reacts
5+ day, 53+ min ago (239+ words) SpaceXAI has disabled a feature in its Grok Build AI coding assistant after researchers discovered that the tool was uploading users’ entire code repositories to Google Cloud without clearly informing them. This issue has drawn attention from thousands of users,…...
Using UMANG app? Your data might be exposed to hackers, tips to stay safe
5+ day, 22+ hour ago (258+ words) Security researchers Akshay C.S. and Viral Vaghela said they found security weaknesses in several UMANG services. They said these issues could expose EPFO Universal Account Numbers (UANs), LPG booking details from at least one major oil company, and Aadhaar numbers that…...
Meta keeps launching AI features that assume consent: Why it keeps backfiring
6+ day, 22+ hour ago (138+ words) Also read: 10 monsoon gadget safety tips everyone should know Also read: Can AI make you a better cricketer? Str8bat co-founder Gagan Daga thinks so And now the Instagram image feature. Same shape, compressed into days instead of years: public accounts opted…...
Anthropic Claude Code under fire over alleged backdoor security risk: What we know
1+ week, 4+ day ago (352+ words) The National Vulnerability Database, a cybersecurity platform linked to China’s Ministry of Industry and Information Technology, shared a security advisory on Wednesday. The agency said that its security researchers had found what they described as a serious security ‘backdoor’ in…...
AI safety tests are flawed: Anthropic finds Claude detects when it’s being evaluated
1+ week, 5+ day ago (204+ words) Also read: Top 10 AI video generators in 2026: Which one should you use? This raises concerns about how the AI industry tests the safety of its products. All red-teaming operations, all alignment benchmarks and all “refusal of harmful requests” tests assume…...
India mulling AI law to tackle deepfakes, cyber attacks and accountability gaps
1+ week, 6+ day ago (409+ words) The Indian government feels that the existing laws were drafted to govern human behaviour and not how the AI will work. While these laws are capable of handling some problems arising from AI, they were never designed for it. Explaining…...
Can AI voice cloning be made safe? Inside a consent-first approach to stopping deepfakes
2+ week, 17+ hour ago (238+ words) Also read: I almost gave up on Google Gemini, but this one simple tweak made it useful Also read: Before Resynced, every Assassin’s Creed game and the best order to play them Awareness about voice scams is present in Mumbai,…...