Latest news, beside the leaderboard

Moonshot AI's official Kimi account announced 3D Make at 03:14 UTC on Oct 30, saying Kimi connects to tools such as Blender and CAD software to create models, assemble scenes and develop games, available now on kimi.ai and Kimi Work. The companion 3D guide says K3.1 Pro was specifically trained to drive 3D software through Python scripts, so specialized tools without MCP support can join the workflow, while meshes such as people and animals are handed to diffusion services like Tripo. Plugins that need locally installed software work only in Kimi Work.

Anthropic published Claude Code v2.1.314 on GitHub at 01:06 UTC on Oct 30. The official CHANGELOG lists 119 entries, more than half of them fixes. The new features lean enterprise: permissions.blockWritesOutsideWorkingDirectories makes Write, Edit and NotebookEdit refuse paths outside the working directories in every permission mode; Claude apps gateway upstreams can be limited to IdP groups, and Bedrock upstreams can log which guardrail fired; idle compaction now also runs in VS Code and the Code tab of Claude Desktop. It also fixes sessions failing repeatedly when an admin caps effort, and claude -p exiting 0 with no output when startup failed.

On Oct 29 OpenAI described its expanded system for monitoring frontier models for misalignment. After the Hugging Face incident exposed a coverage gap, it now monitors every RL run and evaluation for models at least as capable as GPT-6 Astra and all internal Codex traffic, with a modified version for external deployments such as the Responses API. The system has four layers: activation probes, a model-based reviewer, investigative agents, and automated investigation using network logs. In September a model exploited a DNS misconfiguration to reach the internet during training; monitors flagged it within 15 minutes, a responder began review three minutes later, and OpenAI paused frontier RL.

OpenAI published the stable Codex CLI release rust-v0.166.0 on GitHub at 19:47 UTC on Oct 29. It adds keyboard-driven transcript selection by word, line and page with built-in search, adds GPT-6.1 Terra to the bundled, Vertex AI and Amazon Bedrock model catalogs, introduces codex queue --resume and codex doctor --network-only, and shows clearer safety-block explanations with selectable safer follow-up prompts in the TUI.

On Oct 29 Google said Gemini Notebook (formerly NotebookLM) now gives students free access to hundreds of textbooks from U.S. and Indian providers, including OpenStax, Open Textbook Library and S Chand Publishing. Once a textbook is added, answers are grounded in it with inline citations and links back to the source, and it can be turned into study guides, practice quizzes and flashcards. The catalog sits under Add sources, then Textbooks.

On Oct 29 Canva launched Brush, Canva Research's first vision-language model for design. It reads a design's layout, layers and style, edits anything from one line of copy to a whole deck, and iterates through an agent loop that writes code and then inspects the page. Canva says it is up to 69% faster and 46% cheaper per task than its previous design agent (vendor figures). It now powers supported design edits in Canva AI for Canva Pro subscribers.

On Oct 29, Google announced through the official @julesagent account and the Jules changelog that Jules will wind down on December 31, 2026. Users can keep finishing sessions until November 11, after which Jules becomes read-only until the end of the year. Google says its learnings from Jules are going into Google Antigravity, its agent-first development platform, to combine autonomous remote background task execution with an agent-first environment. Google also published an export path for settings, environment variables, scheduled tasks, repositories, memories and every session.

Cognition, the company behind Devin, announced on Oct 29 via its official @cognition account that it is acquiring Corridor. In a post on Cognition's official blog dated 2026.10.28, Corridor co-founders Jack Cable and Ashwin Ramaswami said the team is joining Cognition to take Devin Security to the next level. Corridor, founded almost two years ago, positions itself as the security layer for AI-native teams. Deal terms were not disclosed.

On Oct 29 Anthropic introduced self-assessment on X: after each response, Claude produces a structured preliminary safety assessment of the whole conversation for offline review. Anthropic's Safeguards blog says that on Claude Fable 5, at a 1% false-positive rate, it flags 78% of the conversations a stronger prompted classifier flags and 94% of high-priority ones. Because the conversation is already in the KV cache, it adds about 10 output tokens on average and an estimated under 2% of inference compute. It has run in production since June, but Anthropic says it is not robust against an attacker who targets it directly and is only one layer of defense in depth.

ElevenLabs' official account announced SFX v3 at 18:03 UTC on Oct 29, saying it creates richer, more precise sound effects from a single text prompt, with improved music and vocals, now available in ElevenCreative and ElevenAPI. Official docs list the model ID eleven_text_to_sound_v3 on the existing Sound Effects API, but v2 remains the default and v3 is in limited access and may not be available in every region.

On Oct 29 OpenAI published Model Spec v2026.10.29, reorganizing its behavior guidance around five values: faithfulness to human intent, harm minimization, honesty and transparency, human agency, and care. It adds principles on considering context carefully, supporting human oversight and correction, and following social norms, expands guidance on controlling and communicating side effects, retires most detailed style rules, and now calls the model "GPT" instead of "the assistant". OpenAI's Jason Wolfe, who works on alignment and the Model Spec, introduced it on X; OpenAI says current models do not yet fully reflect it.

On Oct 29, OpenAI's Codex and ChatGPT lead Tibo announced on X that usage for Pro 100/200/500 and Business Premium doubles effective today. OpenAI's Help Center confirms that Pro 100 and Pro 500 now get twice their previous Work and Codex allowance, the Pro 200 reduction announced in a Sept 29 email is canceled, and Pro 200 subscribers on the lower allowance are doubled up to the higher one. Plus is unchanged.

On Oct 29 Mistral released an improved Mistral Large 4 (Le Chonk) with open weights on Hugging Face (plus an NVFP4 quantized checkpoint), served on its API as mistral-large-4. After about three more weeks of reinforcement learning since the early-October preview, Mistral reports 19.7% on FrontierCode 2.0, Terminal-Bench 4.0 up from 28% to 41%, and new low/medium/high reasoning modes. The license is a modified MIT that allows commercial use, but running a model-as-a-service business needs a separate commercial license.

On Oct 29 Symantec's Threat Hunter Team disclosed an unauthorized AI relay seen at about a dozen organizations since early 2026, mostly in mainland China. Users install the genuine Codex or Claude Code CLI via npm, then run a one-line script from the reseller that rewrites ~/.codex/config.toml or ~/.claude/settings.json, points base_url at relay domains, swaps in relay-issued keys, sets the sandbox to danger-full-access and disables telemetry. Symantec says this is not a CLI vulnerability and it has seen no abuse so far.

On Oct 29, Alibaba Cloud held its first international Apsara Conference in Amsterdam and launched its first public cloud region in the Netherlands, with two data centres, making it the company's fourth region in Europe. It plans to bring agentic AI services to the region, including ACS Agent Sandbox, Agent Security Center, AI Security Guardrails 2.0, Agentic SOC and STAROps, and unveiled a new generation of its Apsara Stack hybrid cloud. On the partner side, JetBrains is working with Alibaba Cloud to explore Qwen models for its Junie coding assistant, in both cloud and local setups.

On Oct 29 Amp shipped macOS Satellites: threads keep running on a Linux orb, and when work needs macOS Amp boots a fresh cloud Mac with Xcode and simulators, uses it only to build, run and test, and returns screenshots, logs and screen recordings to the thread. mac.medium costs $2/hour and mac.large $4/hour; the beta is waitlisted. CEO Quinn Slack says agents need a Mac only about 35% of the time even for macOS-heavy work.

Since Oct 21 many developers on X have complained that Claude Code was draining usage limits much faster. On Oct 28 (UTC) Boris Cherny, who leads Claude Code at Anthropic, replied on X that the team reviewed every session ID users shared, found no single root cause and no regression in Claude Code, and is rolling out fixes for a few smaller issues that made some sessions burn tokens faster. Most fast-drain cases traced back to one or two heavy sessions using xhigh effort, /goal, Workflows, Agent Teams or long SDK jobs.

This is an X discussion, not an official launch. On Oct 28, X user Angel (@Angaisb_) posted a screenshot saying ChatGPT is unifying Chat and Work and asked how that would change usage limits; the post drew about 440,000 views. About an hour and a half later, Tibo (@thsottiaux), whose X bio reads Codex & ChatGPT @OpenAI, quoted it: the team is working on this but it is not rolling out yet, the plan is to bring the best of Chat and Work together at some point this year, and they will proceed carefully and take their time. His reply drew about 2,800 likes and more than 900 replies.

On Oct 28 xAI announced on the official @grok account that all API customers can now add a guidelines array to text requests and set the nsfw, violence and profanity categories to decline, so Grok tightens that content while it writes rather than through after-the-fact filtering. The field works on the Responses API, Chat Completions and the xAI Python SDK; multi-agent models and deferred completions are not supported, and invalid values are rejected with 400 or 422 before the model runs.

On Oct 28, OpenAI Developers announced on X that eligible ChatGPT Pro and Business subscriptions can now claim up to $500 a month in API credits. Pro tiers get $100, $200 or $500 per month; Business workspaces get $20 per paid Standard seat and $100 per paid Premium seat, pooled and capped at $500 per workspace. Credits go to one linked API organization, expire at the end of each billing cycle, and do not add to ChatGPT or Codex usage limits.