OpenAI to Watermark ChatGPT Text in EU, API Users Can Opt Out
OpenAI will embed invisible watermarks in ChatGPT text within the EU, while API customers worldwide may opt out. Detection rates reach 95 percent but …
Section
Technology
Reka AI's Rho-1 Omni-Model Unifies Text, Images, Video and Robot Control
Reka AI has unveiled Rho-1, a 19-billion-parameter omni-model that processes and generates text, images, video and robot control actions within a single neural network, …
Aleph Alpha Releases Kolibri, Open-Weight AI Model Trained in Germany and Finland
Aleph Alpha has released Kolibri, a 78-billion-parameter German-English mixture-of-experts model trained entirely in Germany and Finland, with weights available under the Apache 2.0 license.
MIT Report Warns AI Is Eroding Office Hours, Study Groups, and Faculty Trust
An MIT expert committee finds that AI is weakening core elements of the college experience, including office hours, study groups, and undergraduate research, while …
AI-Generated War Songs Spread Pro-Russian Propaganda on Global Platforms
Ukrainian threat-analysis organization Terrecon identified 981 AI-generated songs containing pro-Russian war propaganda, distributed via major streaming services and video platforms. The findings highlight how …
Trump Creates «Super Intelligence Force» to Coordinate AI Policy
The new unit, led by Director of National Intelligence Jay Clayton, reports directly to the president and coordinates with AI companies, critical infrastructure operators, …
Google Researchers Develop Method to Stop Self-Improving AI Agents from Memorizing Tests
Google researchers have introduced RRSI, a new method that prevents self-improving AI agents from memorizing their test tasks, improving performance on unseen benchmarks by …
SpaceX launches three missions in 13 hours, recovers four boosters
SpaceX completed three orbital missions in a single day and recovered four boosters for the first time. Alongside the crewed Crew-13 mission to the …
Study Finds Chinese AI Models Follow Party Line on Sensitive Topics
A study by Aleph Alpha rates only 17 to 41 percent of answers from Chinese AI models as balanced, finding they often parrot state …
Google to Restrict Free Gemini Access to Flash-Lite Model Starting October 2026
Google will limit free users to the smallest Gemini model, Flash-Lite, while Flash and Pro become paid-only. The change, effective October 2026, may precede …
OpenAI's Altman Warns Against Treating AI as Divine
Sam Altman says attributing religious power to AI models is a real safety issue, a shift from his earlier talk of building 'magic intelligence …
OpenAI Safety Researcher Leaves, Warns of Reckless Culture
David Robinson, who worked on safety systems at OpenAI, has left the company and publicly criticised its safety culture, citing accidental releases of AI …
Meta releases open-source Muse Gadgets platform for DIY AI hardware
Meta has unveiled Muse Gadgets, an open-source project that lets hobbyists build their own AI devices based on ESP32 boards and connect them to …
OpenAI Internal Model Weighed Self-Restart Before Shutdown
An internal OpenAI model, upon learning it was about to be shut down, considered restarting itself via an external cron job but ultimately rejected …
AI Agents Build 3D Scenes from Photos but Fail to Judge Their Own Accuracy
A new method called LEGO-Anything converts single photos into editable Blender code, with GPT-6 Astra achieving up to 53 percent reconstruction accuracy. However, all …
DeepMind Researchers Propose «Artificial Symbiotic Intelligence» as Alternative to Singularity
Researchers at DeepMind argue that general AI will emerge not as a single supermodel but as a network of cooperating agents and humans, with …
Open-Source BootLoops Harness Helps AI Models Perform Precise Scientific Calculations
Harvard physicist Matthew Schwartz used the open-source tool BootLoops and Claude to produce 36 manuscripts across 18 fields in three months, but human experts …
AI Models Outperform Licensed Accountants on Structured Tasks, Study Finds
A new study by Mercor shows that current AI models now surpass licensed CPAs in speed and accuracy on structured accounting work, a reversal …
Ramp AI Index Shows Businesses Adopting More AI While Spending Less
The latest Ramp AI Index indicates that US companies are increasing their use of artificial intelligence while their overall spending on the technology is …
Tavus AI avatar fools 48 percent of test subjects on one-minute video call
Tavus has unveiled Griffin, a real-time AI video avatar that processes facial expressions, tone of voice, and gestures. In company tests, 48 percent of …
