Essays on multi-agent delivery, security failures, and the architectural trade-offs that matter when systems leave the demo stage.
Showing 1-15 of 105 posts. Search and filters apply to this page.
OpenAI released GPT-Red for automated model red-teaming, Apple opened its rebuilt Siri to public beta testers, Thinking Machines Lab debuted Inkling, and OpenAI shipped its first Codex hardware while reports detailed a broader device strategy.
Read article →Moonshot AI has launched Kimi K3 through its apps and API, promising full weights for the 2.8-trillion-parameter model on 27 July.
Read article →Anthropic's new Sonnet model brings stronger autonomous coding and tool use to lower-priced tiers, making cost and reliability the next competitive battleground for agentic AI.
Read article →Lumelo is now available for iPhone and iPad: a private family journal for saving the small notes, photos, voice memories, milestones, and reflections that become part of a child's story.
Read article →Apple used WWDC 2026 to relaunch Siri and Apple Intelligence around personal context, Private Cloud Compute, and developer-facing AI tools. The strategy depends on whether users and regulators accept Apple's privacy claims.
Read article →A US export control directive forced Anthropic to suspend access to Claude Fable 5 and Mythos 5 for all users, turning frontier model safeguards into a live product availability issue.
Read article →OpenAI moved closer to public-market scrutiny, Apple used WWDC to reset its AI privacy argument, Microsoft kept explaining its independence from OpenAI, and agent infrastructure spread through coding, cloud procurement, payments and workplace tools.
Read article →OpenAI's new Codex reset banking gives eligible Plus and Pro users a self-service way to recover from rate limits, with referral rewards and 30-day expiry rules attached.
Read article →Anthropic has released Claude Fable 5 with a temporary subscription window, published API pricing and a clear warning that demand may force capacity limits.
Read article →Moonshot AI has released Kimi K2.7 Code, a specialised open-weight coding model with 256K context, multimodal input and benchmark gains on coding and MCP-style agent tasks.
Read article →Anthropic pushed further into public-market, safety and coding-agent territory this week, while OpenAI expanded Codex, Microsoft put agents back at the centre of Windows, and Meta, Apple, Google and Nvidia moved agent systems closer to everyday work.
Read article →Google I/O 2026 centred on Gemini 3.5, Gemini Omni, Search agents, Antigravity, Android 17, intelligent eyewear, Workspace AI and content provenance.
Read article →Anthropic's near-trillion-dollar valuation, Opus 4.8, Robinhood's agent trading move and a fresh wave of enterprise coding-agent deployments set the pace for AI news this week.
Read article →Google used I/O to push Gemini agents deeper into Search, shopping and Workspace, while OpenAI's week centred on healthcare, education, national partnerships, provenance and research claims.
Read article →Gemini Spark turns Google's agent strategy into a consumer and workspace product: persistent cloud agents, MCP integrations and Antigravity as the harness underneath.
Read article →No posts match your search on this page.