Scoop
AI tool news · rumor vs. reality

The AI Wire ●

Rumors tracked. Announcements verified. Updated daily.

Showing Released · show everything

2026-11-02

RELEASED

xAI quietly reroutes Grok Imagine's Quality model to Image 2.0 at low quality — requests succeed, the model answering is different

Since November 2, API requests naming xAI's grok-imagine-image-quality model no longer reach that model: they keep the same request and response shape but are served by grok-imagine-image-2.0 with quality set to low, and the older grok-imagine-image-pro name follows the same chain. Nothing returns an error, and the bill drops by a cent per image at every resolution — the replacement bills $0.04 against the old $0.05 at the standard size, AI Pricing Guru's daily-verified pricing tables show.

xAI did give notice: its migration guide went up with a 60-day window starting September 2 and recommends switching to the new model by name so customers choose the quality they pay for. Even so, this is the third xAI model retirement in five weeks handled by silent redirect rather than an error, one tracker noted, and xAI keeps no dedicated deprecations page and no written general notice policy.

The practical lesson for developers is to log the model that actually answered, not just the one requested — at xAI right now, the two can differ. The redirect lands in the same week xAI switches on a default text watermark for Grok API output, from November 7, which the Wire covered on October 31.

RELEASED

Anthropic puts message threads in public beta: Claude agents stop resending the whole conversation on every turn

Anthropic released message threads in public beta on the Claude Platform on November 2. Instead of resending the full conversation history with every Messages API request, an agent sends the full conversation once, then only the new messages plus the previous response ID, while Anthropic holds the thread in memory between turns.

The gain is latency, not cost. With message threads on by default for most Claude Code users, Anthropic says requests carrying 500,000 to 1 million tokens see median time to first token fall by about 35%, and one reported long-conversation payload shrank from about 800 KB to 1.8 KB. Claude still reads the full conversation on every turn, so token usage, standard rates and rate limits are unchanged.

The limits are built into the design. Threads are held encrypted in memory, never written to disk, and expire five minutes after the last request, after which the client resends the full conversation to start a new one. The beta is available on the native Claude Platform, Claude Platform on AWS, Claude in Microsoft Foundry and Claude Models on Google Cloud, aimed at long-running coding agents, computer-use agents and document- and image-heavy workflows.

RELEASED

Anthropic open-sources FlashProve, a prototype that cryptographically proves which model produced a response

Anthropic's Frontier Red Team, working with collaborators from EPFL, MIT and elsewhere, published FlashProve on November 2: an open-source research prototype for provable inference. Using a zkSNARK — a zero-knowledge proof system — a provider publishes a short commitment to one exact set of model weights without revealing them, then produces a proof for a completed prompt-and-response pair that anyone can check against that commitment.

The test case was large. Anthropic evaluated FlashProve on Meta's Llama 4 Maverick, a roughly 400-billion-parameter mixture-of-experts model in a 4-bit conversion, generating proofs at about 110 tokens per second on a single H100 GPU. A batch of 24 requests took about 223 seconds and produced an 18.78 MB proof per request, while verification took about 17 seconds on a 24-thread CPU with no GPU needed.

The cost is the catch. Proving took roughly 70 to 95 times the GPU time of simply running the model on vLLM, contexts are limited, and Anthropic frames the release — code, Lean proofs and documentation under Apache 2.0 — as a progress update, not a product. A proof also attests only that a specific model produced a response; it says nothing about whether that response was accurate, safe or unbiased.

2026-11-01

RELEASED

Rumor: 'gpt-6.2-astra' is showing up in Codex usage data — multiple accounts, a raw response fragment, and no word from OpenAI

A model OpenAI has never announced is turning up in its own usage accounting. Over the weekend of October 31 to November 1, several Codex users posted screenshots of their usage analytics listing a model called gpt-6.2-astra alongside GPT-6 Astra, the flagship OpenAI launched on September 3. TestingCatalog, reporting on November 2, found the new ID only on the October 31 bar of each breakdown, at shares that vary widely by account — 1.6% of one user's week, 19.8% of another's, and 37.7% of a single day for a third.

The model picker still says GPT-6 Astra, which points to quiet server-side routing — an A/B test or the start of a staged rollout — rather than a launch anyone chose. Watchlist account Token Gremlin (@TokenGremlin) shared a raw response event whose model field reads "gpt-6.2-astra" and called it official; that is strong evidence the ID is live on OpenAI's servers, but it is not an announcement. A separate review by VanData scored the Astra screenshots the strongest evidence of the weekend's leaks, while noting they cannot be traced to a full request.

Nothing official backs the name up. Checks on the morning of November 2 found gpt-6.2-astra in none of OpenAI's public places: not the API model catalog, not the API changelog (whose latest entry is gpt-6-rosalind on October 30), not the Codex changelog, and not the public models.json, which still lists five GPT-6 family slugs. Claims that 6.2 is more powerful, cheaper or faster are unverified. The context makes a successor plausible — OpenAI shelved GPT-6.1 Astra in late September after internal testing raised concerns, so a jump straight to 6.2 would fit a fixed, retrained model — but OpenAI has confirmed no name, price or date. This stays rumor, unresolved.

On November 4, The Information's AI Agenda newsletter (Tiffany Li and Stephanie Palazzolo) reported that new OpenAI models 'could come as soon as Friday,' November 6, citing people with knowledge of the models, and that they 'could include the next generation of its Astra, Sol and cyber focused models' — with the caveat that model timelines could change. The report does not use the name GPT-6.2 Astra, which still rests only on the usage screenshots and a raw response fragment. As of this morning's sweep on November 6, OpenAI has announced nothing, so the rumor stays unresolved — and today is the day to watch.

Midday check, November 6: the Friday window has arrived and, as of the 1:19 p.m. ET sweep, OpenAI has still announced no Astra or Sol model — the only member of the reported trio to ship so far is GPT-6 Cyber, covered here on November 5. A November 5 catalog audit by APIMaster retrieved 242 upstream catalogs and found no GPT-6.2 Sol listing and zero public routes, and OpenAI's model directory and pricing pages still list no 6.2 model. The rumor stays unresolved; if nothing lands by the evening sweep, the reported date will have slipped, exactly as The Information warned it could.

Evening check, November 6: as of the 7:19 p.m. ET sweep, the reported Friday window is effectively closing with no Astra or Sol announcement from OpenAI — GPT-6 Cyber remains the only member of The Information's reported trio to ship. The rumor stays unresolved, and the reported date now looks to have slipped, as the original report warned it could.

Morning check, November 7: still nothing. As of the 7:19 a.m. ET sweep OpenAI has announced no GPT-6.2 Astra or Sol, and no 6.2 model appears in its public catalog or pricing — the reported Friday window has now passed in full. The rumor stays unresolved, with the date slipped.

Morning check, November 11: the story widened to Sol, then thinned again. On November 8, watchlist account @TokenGremlin posted that 'GPT-6.2 Sol is already showing up for some ChatGPT users' — about 277,000 views by November 9 — but the Sol reports rest on output quality and longer Plus usage, with no model ID from an API response, and CellCog's November 9 check found GPT-6.2 in neither OpenAI's API catalog (newest Sol is GPT-6.1 Sol) nor OpenRouter. On November 10, Kingy spotted the phrase 'models 6.2 and later' in OpenAI's Responses API reference — an official documentation mention of the 6.2 version number, though it named no model — and on rechecking later the same day the wording was gone. OpenAI has still announced no GPT-6.2 Astra or Sol, with no name, price or date confirmed: the rumor stays unresolved.

Update, November 12: the traces multiplied overnight but OpenAI still has not confirmed anything. A November 12 radar roundup reports posts showing a gpt-6-astra to gpt-6.2-astra model migration in the Codex repo, GPT-6.2 Sol surfacing on Azure, and both names on Google Cloud quota pages — while checkers note OpenAI's pricing page still lists only gpt-6-astra, gpt-6.1-sol, gpt-6.1-terra and gpt-6-luna. Still a rumor: no model ID, price or announcement from OpenAI.

Radar check, November 12 evening: the rumor heated up without landing. OpenAI's Codex lead Thibault Sottiaux posted on November 11 that 'Thursday is feeling codexy to me. Perhaps also a good time for a banked reset,' and two X accounts claimed GPT-6.2 Sol and GPT-6.2 Astra would arrive today. As of 20:03 UTC on November 12, CellCog found OpenAI's API changelog had no entry newer than November 5, its pricing page and model catalog still listed no 6.2 model, and OpenRouter carried no 6.2 ID. Still unresolved — a tease and launch-day claims, with no official announcement.

Morning check, November 13: still nothing official. Fresh roundups checked on November 12 evening and November 13 agree that OpenAI has not announced any GPT-6.2 model: its model help page lists GPT-6, GPT-6 Pro powered by Astra, GPT-6.1 Sol, GPT-6.1 Terra and the GPT-5.6 family and no 6.2, and checks of the API changelog, pricing table, model catalog and OpenRouter found no 6.2 entry. The Daybreak keynote that some trackers flagged as a possible launch slot came and went with security products only. The rumor stays unresolved.

Evening check, November 13: Thursday passed with no model, and the tease resolved as product updates plus the reset. CellCog's tracker, read at 23:26 UTC on November 13, still finds no OpenAI model with 6.2 in its ID in OpenRouter's catalog and no 6.2 model named in OpenAI's API changelog. Sottiaux posted on November 12 at 21:59 UTC that OpenAI had 'shipped some good updates to dots today and the banked reset has been activated' — the 'codexy Thursday' post produced dots updates and a usage reset, not a 6.2 launch. One unsourced X post on November 13 claims Astra and Sol are likely to come together next week; it cites no source. OpenAI has given no date, and the rumor stays unresolved.

Morning check, November 15: the delay story is now the loudest version of the rumor, and it still has no official backing. A November 13 X post by @kimmonismus, amplified by TLDR on November 14, claims OpenAI delayed GPT-6.2 Astra after being surprised by Claude Fable 5.2, that the model was nearly ready with code-to-video generation, and that Astra and Sol will launch together next week. OrcaRouter's grading of those claims makes the obvious point: OpenAI never announced a 6.2 date, so a model that was never scheduled cannot strictly be late. Checks on November 14 still found no 6.2 row in OpenAI's model catalogue, API changelog (newest entry November 5) or pricing, and no 6.2 row on Artificial Analysis. The rumor stays unresolved.

Midday check, November 15: a syndicated Earth Insider piece dated November 15 claims GPT-6.2 Astra has launched, tying it to the $1.25 trillion valuation figure, but it cites no OpenAI announcement, catalog entry or price. The checkable places still say no: Microsoft's Azure Foundry region-availability page, read this sweep, lists gpt-6-astra, gpt-6-sol, gpt-6-luna, gpt-6.1-sol and gpt-6.1-terra and no 6.2 model, matching the November 14 checks of OpenAI's own catalog, changelog and pricing. The rumor stays unresolved.

Resolved, November 17: it shipped. OpenAI released GPT-6.2 Astra and GPT-6.2 Sol together in ChatGPT Work and Codex, with API access the same day — covered in today's release entry. OpenAI says Sol comes within half a point of Astra on Cognition's FrontierCode 2.0 at about a fifth of the cost, and Astra's standard API prices are 20% lower than GPT-6 Astra's. The model ID that surfaced in Codex usage data on November 1 was the real thing, seventeen days early. This rumor is confirmed.

2026-10-30

RELEASED

OpenAI releases GPT-Rosalind-6, its Astra-based life-sciences model, to approved research organisations only

OpenAI published a safety report on October 30 for GPT-Rosalind-6, a new version of its life-sciences model for biology, drug discovery and translational medicine, and added it to the API changelog the same day. Access is tightly gated: GPT-6 Rosalind Research is available to approved research organisations through ChatGPT Work and Codex, API access needs a separate approval, and buying a subscription alone does not provide it. Applicants face a trusted-access review of their proposed research, security controls and governance, and in Enterprise workspaces the model starts with a two-week administrator preview, off by default.

The model uses the same underlying weights as GPT-6 Astra, with no additional training for this release. OpenAI says it refuses fewer legitimate biological questions that could also have harmful uses, while still refusing biological weaponisation. It is classified High capability in the biological and chemical domain and, relying on GPT-6 Astra's assessments, Critical in cybersecurity; none of its four Critical biological capability evaluations reached the relevant threshold.

On OpenAI's own benchmarks, at the highest reasoning budget tested, GPT-Rosalind-6 scored 49.6% on a medicinal-chemistry evaluation against 28.9% for GPT-Rosalind-5.5, 37.7% against 16.7% on genomics and quantitative biology, and 61.3% against 44.0% on an applied life-sciences benchmark built from tasks contributed by working scientists. OpenAI notes one biological-data benchmark used a configuration different from the production system, and the scores do not show the model can diagnose patients or deliver a successful medicine.

Pricing update, November 2: OpenAI's API pricing page, checked on November 1, lists gpt-6-rosalind at $15 per million input tokens, $1.50 cached and $75 output for prompts up to 272K input tokens — rising to $30, $3 and $112.50 above that — against GPT-6 Astra's $10 and $50. That is a 1.5x premium for, on OpenAI's own October 30 system-card addendum, the same Astra weights with no added training data and no additional post-training: what trusted-access customers are paying for is a looser refusal boundary on dual-use biology, paired with stricter cyber refusals. The Astra for Law configuration, by contrast, still carries no premium, at the standard $10 and $50.

RELEASED

Google moves Gemini Computer Use into general availability for developers

Google has moved its Gemini Computer Use tool into general availability, giving developers a production release of the software that lets AI agents interact with browser, mobile and desktop screens. The change appears in the October 30 Gemini API changelog and applies to both the Interactions and GenerateContent APIs. It is a change in release status, not the technology's first introduction — the changelog records earlier preview versions.

Computer Use lets a model analyze screenshots and propose actions such as clicking, scrolling and typing; developers must supply the software that carries out those actions and sends updated screenshots back. Google lists repetitive data entry, website form filling, web application testing and research across websites among the uses. Its documentation also describes configurable safety policies and optional screenshot scanning to detect hidden adversarial instructions, with an action system that can require user confirmation or block an action outright.

General availability is not a clean bill of health: Google warns Computer Use may contain errors and security vulnerabilities, recommends close supervision for important tasks, and advises against using it for critical decisions, sensitive data or actions where serious errors cannot be corrected. It is also a developer tool — not an announcement that Gemini chat users now have an autonomous computer operator.

RELEASED

Claude Code now refuses to recreate specific existing games under a new server-side rule

Anthropic has quietly added a server-side rule to Claude that blocks users from recreating a specific existing game — even if it is renamed or called original. The rule, delivered under a flag called 'tengu_heron_brook,' first appeared in Claude Opus 5.5's changelog on October 27 and was confirmed in reporting on October 30; there is no setting to turn it off.

Original work is still allowed. The rule expressly permits building original games, games in a genre, and games from the user's own designs and assets — the line is drawn at recreating a particular existing title.

The change lands after weeks of users pairing Claude with reverse-engineering tools to decompile and rebuild existing games, and users are now reporting that related new safety classifiers seem to be limiting other work too, from security research to published coding books. Anthropic has not published a detailed explanation of the rule's scope.

2026-10-29

RELEASED

Mistral publicly releases Large 4, its most capable open model for cybersecurity work

Mistral released its Large 4 model to the public on Thursday, October 29, Reuters reported. Unlike leading AI systems such as Google's Gemini, which are mainly accessed through online services, Large 4 can be downloaded and run independently by anyone, on their own systems.

The released version is an upgrade from the one Mistral unveiled three weeks ago, following testing with governments and corporate users that helped improve its performance. Mistral described it as its most capable openly available model for cybersecurity use — designed to help organisations identify and fix vulnerabilities — competing in a field dominated by Chinese AI models.

"I'm not comfortable living in a world in which open weights do not exist," Mistral Vice President of Science Pierre Stock said. Once the model is out, he added, "it's impossible to contain access" — that protects users from being cut off, but it also means anyone can use it.

RELEASED

Meta launches Waddle World, an AI pet that turns your day into illustrated diary entries

Meta has launched Waddle World, an AI virtual-pet app now available on iOS and Android in the Philippines, with wider availability unconfirmed. Users hatch and care for a pixelated companion in a 16-bit style, feeding it, playing with it and watching its mood change.

The twist is conversation: users talk or type about their day, and the pet — which remembers what they share — turns those conversations into illustrated journal entries. The store listing describes companions with moods such as happiness, hunger, sleepiness and boredom, learning more about the user's routines over time, and specifies users must be at least 18.

Waddle World is one of several experimental consumer AI apps Meta has shipped in recent months, alongside Style Sketch, Pocket and Vibes. Meta has not confirmed any wider release or said whether the app connects to its broader Muse AI companion work.

RELEASED

Meta quietly launches Style Sketch, an AI fashion app that builds outfits from clothes you already own

Meta has quietly released Style Sketch, a free AI fashion app now available on iOS and Android in Indonesia, Singapore and Thailand. The app was first spotted by social media consultant Matt Navarra on October 28, and Meta did not respond to TechCrunch's request for comment or say whether a wider launch is planned.

Style Sketch builds a virtual closet from photos a user has already posted to Instagram, or from items added manually by photo, video or camera-roll upload. It then suggests outfit combinations matched to the user's style, with virtual try-on suggestions.

The launch is part of a run of small standalone AI apps from Meta in recent months, alongside Vibes for AI video, Pocket for making games, and the AI pet app Waddle World. TechCrunch notes that digital-closet rivals such as Alta and tools in Google Photos already exist, so Meta's edge is the Instagram connection rather than the idea itself.

RELEASED

Rumor: Grok Bot 1.0 installers have appeared on Cursor's download server, but SpaceXAI hasn't announced it

Installer files labelled Grok Bot 1.0.0 have appeared on Cursor's download server, grokbot.guru reported on October 29. The site checked the Mac installer for Apple silicon on Thursday morning with a HEAD request — which reads a file's details without downloading it — and the server returned the 155.6 MB .dmg, dated October 27 at 6:14 p.m. CT. A check of the Windows x64 installer the same morning also returned a working response.

Nothing official matches it yet. SpaceXAI hasn't announced 1.0.0, the official Grok Bot changelog has no entry for it, and the newest official release there is still 0.78.1, released October 27, which added shopping cards and subscription templates. What the 1.0.0 version changes, and when it will be released, are not known. Kevin Logan posted on X on October 27 that "Grok Bot 1.0 Stable just showed up on the download server," and asked the next day whether anyone had seen it pushed out to users yet.

grokbot.guru advises waiting for the update to arrive through the app's Updates pane rather than installing 1.0.0 from a direct file link before SpaceXAI releases it. This stays a rumor until the changelog or an announcement lands.

An October 30 follow-up from TestingCatalog adds colour but no confirmation: it reports SpaceXAI is preparing Grok Bot 1.0.0 as the first stable release, with the build already downloadable ahead of an official rollout, and spots features in development alongside it — a more direct Microsoft Teams integration, a dedicated Tasks tab, a Texting tab in Settings with permissions to read and send text messages, and early signs of Discord support. None of those features has a confirmed timeline, and there is still no official 1.0.0 announcement or changelog entry, so the rumor stays unresolved.

An October 31 TestingCatalog daily brief goes a step further, describing Grok Bot 1.0 as 'the first stable build' that is out for download, and adds two more sightings: Grok Bots rolling out to X Chat, so users can message their bots like any contact without leaving the X app, and per-bot voice and voice-language settings. The build is plainly circulating — but there is still no announcement or changelog entry from SpaceXAI itself, so the rumor stays unresolved.

Update, November 11: Confirmed. SpaceXAI has officially released Grok Bot 1.0 — its launch post on x.ai is dated November 2 and the public launch push landed on November 11 — with shared Team Bots in Slack, Microsoft Teams and video calls. See the November 11 Grok Bot 1.0 launch entry. This rumor is now confirmed.

RELEASED

Gemini app can now generate, preview and edit Docs, PDFs, Sheets and slides without leaving the chat

Google added file generation, preview and editing inside the Gemini app on October 29. Users can now work on PDFs, Google Docs, Google Sheets, slide decks, Markdown files, TXT files and code without leaving the chat — expanding a file to full screen to type edits directly, or highlighting text to ask Gemini a question or have it refine the file, such as turning a highlighted table into a graph.

The update began rolling out to all Gemini users, in all languages where the Gemini app is available. PDF preview and editing work on both web and mobile; every other format is web-only for now.

Gemini could already create and download files earlier this year, but editing what it generated meant leaving the app. Google's announcement does not mention pricing tiers, usage limits or admin controls for the new editing features.

A November 1 follow-up from Notebookcheck, testing with a free Google account in Germany, confirms the practical detail Google left unstated: no subscription is required — free accounts get the editor too, and it was already live in the web version. On phones the limitation is sharper than the launch coverage suggested: the Android and iOS apps can edit PDFs only, so reworking a presentation or spreadsheet still means switching to a computer.

RELEASED

Gemini lets you share your personal avatar, so friends can star together in AI images and videos

Google is making Gemini's personal avatars social. Users can now share their avatar with friends, tag each other in prompts, and create AI images and videos featuring everyone's digital likeness together, with character and voice consistency maintained across the content.

A Gemini avatar is built from recordings of your face and voice, made by scanning a QR code from the web app and following the phone setup. Once someone shares their avatar with you, you can tag them in a prompt the same way you tag your own.

Sharing is controlled and revocable: you invite people by Gmail address from the Avatar settings, they can accept or decline, and you can add or remove anyone's access at any time. The feature requires users to be at least 18 with a personal Google account, supports English, and is not available in the European Economic Area, Switzerland or the UK.

RELEASED

Anthropic details Claude self-assessment: after each reply, Claude judges its own conversation for misuse

Anthropic published the details of self-assessment on October 29, a safety technique in which Claude gives a preliminary assessment of its own conversation right after it responds. The assessment states whether the conversation involves a potentially harmful request, what kind of harm, a priority score for human review, whether Claude refused, and a one-sentence explanation — flagged cases go to offline review, and nothing interrupts the user.

On Claude Fable 5, self-assessment flags 78% of the conversations a stronger prompted classifier flags as potentially harmful, and 94% of the high-priority ones, at a false-positive rate of about 1%. Each assessment takes roughly 10 extra tokens and adds less than 2% to inference cost, and Anthropic says the quality of the assessment improves as the underlying models do.

Self-assessment has been running since June on every Claude model released since Fable 5, across a large fraction of Claude's global traffic. Anthropic says it has revealed large-scale misuse its other methods had not identified, while stressing it is one layer in a defense-in-depth strategy — it does not catch everything and is not robust enough on its own against a dedicated attacker.

2026-10-28

RELEASED

SpaceXAI gives Grok developers custom moderation controls for NSFW, violence and profanity

SpaceXAI's developer documentation and October release notes describe a new 'Custom moderation controls' option for Grok: a guidelines field on text requests through the Chat Completions and Responses APIs and the xAI Python SDK. It covers three categories — nsfw, violence and profanity — with, for now, a single behaviour, decline.

With nsfw set, Grok avoids sexual innuendo or sexualised conversation but still answers educational and clinical questions; with violence set, it avoids graphic depictions while still engaging with history and news; and with profanity set, it writes without profanity. Any category left out keeps the model's default behaviour.

The settings apply only to the request that carries them, so an app has to send them with every message, though the Python SDK can store them on a chat. They do not work with multi-agent models or deferred completions, and SpaceXAI has not announced the feature beyond its documentation or said how it tests that Grok follows the settings.