Total 1,098
Wu Yonghui, a former research vice president at Google DeepMind, reportedly joined ByteDance on 17 February 2025 to lead foundational research for the Seed large-model team, reporting directly to CEO Liang Rubo.
On 17 February 2025 StepFun published the Step-Audio technical report and released code and weights for its open speech models, jointly with Geely.
On 17 February 2025 StepFun and Geely jointly open-sourced Step-Video-T2V, a 30-billion-parameter video model, and Step-Audio, a speech model family.
On 17 February 2025 xAI unveiled Grok 3 and Grok 3 mini, with a Think button for step-by-step reasoning and the DeepSearch research agent. Free users received Grok 3 on 20 February.
On 14 February 2025, the UK minister for science, innovation and technology announced at the Munich Security Conference that the AI Safety Institute would become the AI Security Institute, focusing on security-related risks and adding a criminal misuse team.
On 14 February 2025 Perplexity launched Deep Research, which runs dozens of searches, reads hundreds of sources and writes a full report. The company reported 21.1% on Humanity's Last Exam.
On 13 February 2025 Tencent's Yuanbao app added DeepSeek-R1 beside Tencent's own Hunyuan models, and within days it rose to second place on China's App Store free chart.
In February 2025 a Delaware federal judge held that Ross Intelligence's use of Westlaw headnotes to train its legal AI tool infringed Thomson Reuters' copyright, and rejected its fair use defence — described as the first substantive AI copyright ruling.
France and India co-hosted the Paris AI Action Summit on 10–11 February 2025, where 60 countries and international organisations signed a statement on inclusive, sustainable AI. The United States and the United Kingdom did not sign.
On 6 February 2025, Mistral AI released iOS and Android apps for its Le Chat assistant, alongside a $14.99 Pro tier, a Team plan and an Enterprise preview, adding fast answers, code execution and image generation.
On 4 February 2025, Figure ended its collaboration agreement with OpenAI. Brett Adcock cited integration problems and said Figure had built end-to-end robot AI in-house, while pointing to BMW and a new large customer.
On 2 February 2025 the first binding parts of the EU AI Act took effect: bans on certain AI practices and requirements on AI literacy. Guidelines on the banned practices followed two days later.
On 2 February 2025 OpenAI launched deep research in ChatGPT, an agent built on a version of the o3 reasoning model that produced research reports in roughly 3 to 30 minutes and was offered to Pro users first.
On 31 January 2025 OpenAI released o3-mini, the first reasoning model it made available to free ChatGPT users, alongside an API release priced per million tokens.
On 30 January 2025 Mistral AI released Mistral Small 3, a 24-billion-parameter model under the Apache 2.0 licence, scoring over 81% on MMLU and running at about 150 tokens per second.
The first International AI Safety Report, published on 29 January 2025, was led by Yoshua Bengio and written by over 100 experts with backing from 30 countries; it sets out the science of general-purpose AI capabilities, risks and mitigations without making policy recommendations.
On 28 January 2025 Alibaba unveiled Qwen2.5-Max and said its instruct model beat DeepSeek V3 on several benchmarks. The model's weights were not released, and access went through Qwen Chat and the Alibaba Cloud API.
On 27 January 2025 DeepSeek's mobile app passed ChatGPT as the most downloaded free app on Apple's U.S. App Store. The same day Nvidia shares fell 17 percent, the biggest one-day loss in U.S. history for any company.
On 27 January 2025 DeepSeek released Janus-Pro, a single model that handles both image understanding and text-to-image generation, in 1B and 7B sizes.
On 24 January 2025 Baichuan released Baichuan-M1-preview and open-sourced Baichuan-M1-14B, a smaller model it described as the first open-source LLM trained from scratch for medicine.
On 23 January 2025 OpenAI released Operator, a research preview agent that used its own browser to fill forms, place orders and schedule appointments for ChatGPT Pro users in the US. It combined GPT-4o vision with reinforcement learning and introduced confirmation and takeover safeguards.
