45 articles
OpenAI has released GeneBench-Pro, a 129-problem benchmark that tests whether AI agents can exercise the judgment-intensive analytical reasoning required for real-world computational biology research.
Read more →
Google has released Gemini Omni Flash in public preview, a multimodal AI model that generates and edits short videos through natural-language conversation, available via the Gemini API at $0.10 per second of output.
Read more →Chinese technology company Meituan has released LongCat-2.0, a 1.6-trillion-parameter coding model trained entirely on domestic Chinese chips, under the MIT License — the first frontier-class open-source model built without Western semiconductors.
Read more →GLM-5.2, released on June 16 by Beijing startup Z.ai, sits alongside Claude Opus 4.8 and GPT-5.5 on coding benchmarks while carrying an API price roughly one-sixth that of closed US models.
Read more →A 40-member UN scientific panel released the first independent global AI assessment on July 1, warning that current safeguards cannot keep pace with the technology and that benefits risk widening global inequality.
Read more →Anthropic's new Claude Science workbench integrates over 60 scientific databases and tools into a unified AI environment for life sciences researchers, launching in beta on June 30, 2026.
Read more →Google has pushed its flagship Gemini 3.5 Pro model from June to July 2026, while a Nobel Prize winner and three other senior researchers departed for OpenAI and Anthropic in the same week.
Read more →Anthropic released Claude Science in beta on June 30, an integrated environment connecting AI to 60+ scientific databases and lab compute for biomedical and life sciences research.
Read more →OpenAI on June 22 expanded its Daybreak security program with GPT-5.5-Cyber, a model designed to detect and patch code vulnerabilities, and launched Patch the Planet to harden over 30 major open-source projects.
Read more →Anthropic's June 2026 Economic Index links real Claude session data to a direct survey of 9,700 users, finding roughly half say AI already handles 50 percent or more of their work tasks — and more than one-third expect automation to cover most of their job within a year.
Read more →Alibaba's Qwen team released a language world model that learns to predict agent environment states across seven domains, offering a new approach to training AI agents without live systems.
Read more →Liquid AI released LFM2.5-230M, a 230-million-parameter open-weight model that runs AI agent tasks at 213 tokens per second on a phone CPU and 42 tokens per second on a Raspberry Pi — without a cloud connection.
Read more →Jonas Adler and Alexander Pritzel — both key figures in Gemini development and AlphaFold — are leaving Google DeepMind for Anthropic, extending a six-day talent exodus that erased roughly $270 billion from Alphabet's market value.
Read more →New data from SignalFire tracking millions of tech workers shows engineers are the most resilient job category in the AI era, even as recent graduates and non-technical roles absorbed the steepest cuts.
Read more →ByteDance launched Doubao Seed 2.1 Pro and Turbo on June 23, claiming top scores on coding and agent benchmarks and parity with GPT-5.5 and Claude Opus 4.7.
Read more →Google has postponed its flagship Gemini 3.5 Pro to July, slipping past the June deadline CEO Sundar Pichai set at Google I/O — the second consecutive I/O commitment to miss its target.
Read more →A multi-institution index of 30 deployed AI agents finds that 25 lack published internal safety results and 23 have no third-party testing, raising urgent questions as agentic systems take on increasingly consequential tasks.
Read more →Chinese lab Zhipu AI has open-sourced GLM-5.2, a 753-billion-parameter model that outperforms GPT-5.5 on software engineering benchmarks while costing roughly one-sixth as much.
Read more →Alibaba's Qwen team has released Qwen-AgentWorld, a family of open-source language world models that predict how digital environments respond to AI agent actions, with its 397B variant outscoring GPT-5.4 on the new AgentWorldBench benchmark.
Read more →Google DeepMind is investing $75 million in indie studio A24 to co-develop AI filmmaking tools — creators keep full control, and Google gets no access to A24's content library.
Read more →OpenAI unveiled three new models — Sol, Terra, and Luna — under the GPT-5.6 banner, with access currently restricted to a small group of trusted partners cleared by the U.S. government.
Read more →