On September 3, 2026, researchers at Google DeepMind published a case study on 100 AI agents tasked with solving mathematical conjectures. A flaw in the evaluation system spread through the collective, while other agents tried…
Muse can buy, book, and negotiate for you: Meta moves into agentic AI
Launched by Meta on September 8, 2026, Muse is presented as a personal agent capable of browsing the web, filling out forms, preparing purchases, booking certain services, negotiating, and continuing tasks in the background. What…
Gemini 3.8 Flash and Flash Cyber Take On Cyber Defense
Presented by Anthropic on September 1, 2026, Claude Fable 5.1 and Claude Mythos 5.1 are built on the same model, but they do not unlock the same sensitive uses. Fable is widely available with strengthened…
Claude Fable 5.1 and Mythos 5.1 Put to the Safeguards Test
Presented by Anthropic on September 1, 2026, Claude Fable 5.1 and Claude Mythos 5.1 are built on the same model, but they do not unlock the same sensitive uses. Fable is widely available with strengthened…
GPT-6 Astra: what its agentic capabilities actually change
Launched by OpenAI on September 3, 2026, GPT-6 Astra combines reasoning, tool use and control of digital interfaces to carry out tasks that go beyond simply producing text. The published results show significant progress in…
Qwen3.8-Max: what Alibaba’s model really changes for coding and AI agents
Launched by Alibaba on August 3, 2026, Qwen3.8-Max combines a Mixture-of-Experts architecture with 2.4 trillion parameters, of which 95 billion are activated, with an advertised context window of one million tokens and multimodal and agentic…
GLM-5.3: what its gains in coding and cybersecurity really allow us to claim
Unveiled by Z.ai on August 14, 2026, GLM-5.3 keeps the same base model as GLM-5.2, but improves through post-training focused on software engineering, long agentic tasks and vulnerability research. Its published results are impressive on…
Kimi K3: what you need to know about the 2.8-trillion-parameter open-source model
For a long time, the global artificial intelligence race seemed dominated by American labs. OpenAI, Anthropic, Google, and xAI held the top spots in international rankings, while Chinese models struggled to compete on the most…
SpaceXAI launches Grok 4.5: an AI designed to code, reason, and automate business tasks
No sooner have companies begun integrating GPT-5.5, Claude Sonnet 5, or GLM-5.2 than a new competitor has already entered the race. SpaceXAI, Elon Musk’s artificial intelligence company, has just officially unveiled Grok 4.5, a next-generation…
Anthropic Makes a Big Splash: Claude Sonnet 5 Offers Nearly the Same Power as Opus at Half the Price
Agent-based artificial intelligence continues to evolve at a rapid pace. After reserving its most advanced capabilities for Claude Opus 4.8 for several months, Anthropic is now shifting its strategy with the launch of Claude Sonnet…