Tencent's new Hunyuan-Large-Vision model now leads all Chinese entries on the LMArena Vision Leaderboard, ranking just behind GPT-5 and Gemini 2.5 Pro.

Google DeepMind has claimed its first gold medal at the International Mathematical Olympiad (IMO) with an "advanced version" of its Gemini model running in Deep Think mode.

OpenAI CEO Sam Altman sees a new appreciation for human-generated content on the horizon, even as AI tools reshape the internet.

According to Character.ai CEO Karandeep Anand, users spend an average of 80 minutes a day chatting with AI-generated fictional characters.

Tencent's X-Omni team shows how reinforcement learning can fix the usual weaknesses of hybrid image AI systems. The model excels at rendering long texts in images and sometimes sets new performance ...

OpenAI is rolling out fixes for early problems with GPT-5, promising short-term upgrades to capacity, quality, and the user interface.

Israeli researchers have discovered that Google's Gemini assistant can be tricked into leaking sensitive data or controlling physical devices through hidden instructions buried in something as ...

Amazon told Sky News it enforces content guidelines using both automated and manual review. Books that break the rules—whether AI-generated or not—are removed. The company says it is constantly ...

Microsoft is bringing OpenAI's GPT-5 to its entire Copilot lineup on Windows, Mac, and mobile devices. In Microsoft 365 Copilot, licensed users can use GPT-5 for more advanced tasks. The Microsoft ...

In the ARC-AGI-2 benchmark, which is designed to measure a language model's general reasoning skills, GPT-5 (High) scored 9.9 percent at a cost of $0.73 per task, according to ARC Prize.

According to a Gartner survey of 3,000 jobseekers, 6 percent admitted to cheating during interviews - either by posing as someone else or having someone impersonate them. Gartner expects the rise of ...

A new benchmark called OdysseyBench puts AI agents through realistic, multi-day office workflows, and the results are surprising: OpenAI's older o3 model consistently outperforms the newer GPT-5 on ...