The archive
137 stories from 60 issues. Filter by industry, type of news, date or company, and choose how much to read.
Industry
25 stories
- 18 NovNewClaude's Fable 5.2 beat OpenAI's GPT-6.2 Astra at real office tasks
Why it matters Benchmark ties hide big gaps: Fable 5.2 scored 98% on data center analysis against Astra's 65%, so test models on your own tasks.
Impact Benchmark ties hide gaps, so teams must test models on their own tasks.
What happened Epoch AI graded six AI systems on tasks drawn from its own work. Claude Fable 5.2 beat GPT-6.2 Astra by 17 points, even though the two tie on a general benchmark. Astra averaged 65%.
- 18 NovNewSenseNova's small AI matches a model three times larger on analysis work
Why it matters Teams paying to run large models for analysis and coding agents can test a model a third the size with similar scores.
Impact Teams can run a model a third the size for analysis, cutting costs.
What happened SenseTime has put out a compact AI that reads text and images, with weights and code available to download. It reuses some of its layers twice, and scores 65.31 on AIDABench against 65.01 for a model three times its size.
- 18 NovNew
Launch
Google's Gemini app now speaks in 66 local voices, including Hindi and KoreanWhy it matters Teams serving customers in India, Brazil, Indonesia or Korea can now test Gemini voices that sound local, not generic.
Impact Companies serving customers in several countries can test local-sounding Gemini voices instead of generic ones.
What happened Google added 66 regional voices to its Gemini app on Android and iOS, covering U.S., U.K. and Indian English, Hindi, Brazilian Portuguese, Indonesian and Korean. The chosen voice applies across the whole app, including live conversations and read-aloud replies.
- 17 NovNew
Launch
Google's AdMob now has an AI helper that fixes ad problems for app makersWhy it matters App publishers on AdMob can hand setup fixes and ad-code checks to AI, with every change still needing their approval.
Impact App publishers can delegate ad troubleshooting and code checks to an AI assistant.
What happened Google is adding Gemini-powered helpers to AdMob: a chat assistant called Ask AdMob for troubleshooting and earnings ideas, plus ready-made agent playbooks for writing and checking ad code. Testing opens over the next few months, and a connector linking developer tools to AdMob data is still in development.
- 15 NovNew
Launch
OpenAI lets ChatGPT's AI assistants pay for purchases with user approvalWhy it matters If your staff use ChatGPT Work or Codex, they can now buy through AI agents, so decide who approves and which cards.
Impact AI agents can now complete purchases, so firms must set approvers and cards.
What happened OpenAI launched ChatGPT Wallet, which lets AI assistants inside ChatGPT Work, Codex and dots pay for things. Each purchase needs the user's sign-off, and the user's existing card protections still apply.
- 12 NovNew
Partnership
Google and Adobe link PDF editing and design tools into GeminiWhy it matters If your teams use Google Workspace, PDF edits and Adobe creative work can happen inside Drive and Gemini, cutting app switching.
Impact PDF editing and design work move inside Drive, Gmail and Gemini, reducing app switching.
What happened Google and Adobe expanded their partnership at Adobe MAX. Acrobat will be reachable from Google Drive and Gmail, Adobe's creative tools are coming to Gemini and AI Mode in Search, and Adobe's CX Enterprise Coworker agent arrives in Gemini Enterprise.
- 10 NovNew
Launch
OpenAI's ChatGPT Enterprise now goes on your Amazon cloud billWhy it matters If you have an AWS agreement, buying ChatGPT Enterprise no longer needs a separate contract, so procurement and approval steps shrink.
Impact Teams buy ChatGPT Enterprise on existing AWS billing, skipping separate contracts.
What happened OpenAI's ChatGPT Enterprise can now be bought through AWS Marketplace. Companies can pay for it on the Amazon cloud bill they already have, rather than setting up a separate contract.
- 10 NovNew
Launch
Google lets your developers try map-powered apps without a credit cardWhy it matters Teams can prototype location-based apps in one place without a credit card, so a Maps proof of concept costs nothing to start.
Impact Developers prototype Maps-powered apps without a credit card or Cloud setup.
What happened Google AI Studio now lets developers add Google Maps data to apps they build there, with no trip to Google Cloud to set up payment or keys. Users can start without a credit card, or link a Cloud project they already have.
- 6 NovNew
Deployment
Merge uses OpenAI's AI for code checks, marketing, sales updates and dinner planningWhy it matters Merge applies OpenAI tools to engineering, marketing, sales and even event planning, a template for spreading one AI vendor across departments.
Impact One AI vendor now serves engineering, marketing and sales, a model for cross-department rollout.
What happened Merge's teams run GPT-6 Astra, ChatGPT and Codex across their daily work, from checking product code to supporting marketing. Staff also lean on AI to organise executive dinners and get sales updates ready.
- 6 NovNew
Launch
Google's Gemini turns a hand-drawn sketch into a working app prototypeWhy it matters Non-engineers on your team can turn a rough sketch into a working prototype in one sitting, before anyone books developer time.
Impact Non-engineers can turn sketches into working prototypes before developers get involved.
What happened Google's Gemini can turn an uploaded image of a hand-drawn app layout, habit tracker or game board into working code. Users pick the Canvas tool, ask for the app to be built, then preview it and tweak colors, fonts or features as they go.
- 5 NovNew
Funding
OpenAI buys a company to improve its AI coding helper, CodexWhy it matters Teams using OpenAI's Codex get more built-in help with code review and comprehension, so evaluate it against your current coding tools.
Impact Coding teams get built-in code review and explanation, worth comparing against current tools.
What happened OpenAI is buying Pierre Computer Company to strengthen Codex, its coding product, with tools for writing, explaining and checking software.
- 4 NovNew
Programme
Google's charity arm funds five groups building AI tools for teachersWhy it matters Education-technology vendors now face grant-funded rivals building open lesson-planning tools for teachers, so paid products must justify their price.
Impact Paid classroom tools face grant-funded open rivals, so pricing needs stronger justification.
What happened Google.org is giving more than $9.5M in grants to five groups working on AI for classrooms: CUNY, Fab AI, Krause Center for Innovation, Teaching Lab and 1EdTech Global. Fab AI is working with Google DeepMind and the Gates Foundation on open lesson-planning and early-reading tools.
- 3 NovNewOpenAI built its own code storage system to cope with AI-written software
Impact Code hosting must be rebuilt to handle the volume of AI-written software.
What happened OpenAI built Pando, a backend that works with Git and is designed to handle the volume of code development that its Codex tool produces.
- 2 NovNewMicrosoft's small AI can operate a computer by clicking and typing commands
Why it matters An openly licensed AI this small can now operate a computer, so test it before paying for a bigger one.
Impact A small, cheap model now handles computer-operating agents, cutting costs.
What happened Microsoft released Mage-VL-Cowork, a 4B-parameter AI that operates a computer by clicking on screenshots and by running code in a terminal. It scores 66.0% on OSWorld-Verified with a 50-step budget, the best result among 4B models.
- 2 NovNewOpenAI is adding earlier safety checks to ChatGPT and will publish safety reports
Why it matters If your product or staff use ChatGPT, expect stricter intervention in some conversations, and a public OpenAI safety report to cite in risk reviews.
Impact Chatbot builders face expectations of earlier interventions and public safety reporting.
What happened OpenAI is adding safeguards in ChatGPT aimed at preventing violence, stepping in earlier when conversations raise concern. It has also committed to publishing public safety reports.
- 1 NovNewWebsite owners can now tell OpenAI whether their content may train its AI
Why it matters If you publish content, one robots.txt entry can now state your AI training preference to OpenAI and other participating providers.
Impact Publishers can state AI training preferences to OpenAI through one robots.txt entry.
What happened OpenAI will respect a shared robots.txt signal that lets website operators state whether their content may be used for AI training. The signal is meant to work across participating AI providers.
- 30 OctNewOpenAI rewrites the rules for how its AI is meant to behave
Why it matters If you build on OpenAI's AI, its behaviour rules just changed, so re-test your customer-facing prompts and workflows against the new version.
Impact Teams building on OpenAI models get a clearer picture of intended model behavior.
What happened OpenAI revised its Model Spec, the document describing how its AI systems are meant to behave, and organised it around five core values. Among them are good judgment, keeping people in control, and staying true to what users actually intend.
- 30 OctNewFree AI from Z.ai can hack unpatched Firefox for about $1,000
Why it matters Attackers can now turn a published fix into a break-in for about $1,000, so any delay in installing browser updates is a real exposure.
Impact Cheap AI-built exploits from published fixes make delayed browser patching a real exposure.
What happened Epoch AI reports that Z.ai's openly available GLM 5.3 turned published-but-undelivered Firefox security fixes into a working attack that stole a file from a visitor's computer with no clicks. Each Firefox build from mid-September to late October fell to it, at roughly $1,000 per build; Mozilla's 159.0 update on October 27 closes one flaw the attack relied on.
- 29 OctNew
Partnership
ServiceNow customers get OpenAI's AI assistants built into their work processesWhy it matters ServiceNow customers can reach OpenAI's agents from inside their existing workflows, and ChatGPT users can trigger ServiceNow processes without switching tools.
Impact Teams can run OpenAI agents inside ServiceNow and trigger its workflows from ChatGPT.
What happened OpenAI and ServiceNow are widening their partnership. OpenAI models and agents will run inside ServiceNow, and ServiceNow workflows will become available inside ChatGPT and Codex.
- 28 OctNew
Deployment
Retell's phone-call AI business earns $1.6M per employee using OpenAI toolsWhy it matters Retell reports $1.6M annual revenue per employee using OpenAI's Codex, a productivity figure to measure your own team against.
Impact Voice-agent builders now have a benchmark leader and a revenue-per-employee target.
What happened Retell, a voice-agent company handling 100M calls a month, says OpenAI's GPT-6 Astra scored highest on its internal voice benchmark. It also says OpenAI's Codex helps it earn $1.6M in annual revenue per employee.
- 27 OctNew
Regulation
UK regulator is drafting rules on how Google shows people their search choicesImpact Search advertisers and rival search providers face new rules on how Google presents choices.
What happened The UK Competition and Markets Authority ran a consultation from 23 September to 9 October 2026 on updated rules for how Google presents search choices to users. Google was designated in October 2025 as having strategic market status in general search and search advertising; responses are still being processed.
- 26 OctNewNanbeige's new text-writing AI is free to build into your products
Why it matters Teams testing mid-sized AI for coding and office tasks have another contender to benchmark, with a 3B smaller version planned.
Impact Another openly licensed mid-sized text model is available to build into products.
What happened Nanbeige has published Nanbeige4.3-33B, a 33B-parameter text-generation model, on Hugging Face under the Apache-2.0 licence. It has 372 likes and no recorded downloads.
- 25 OctNewOpenAI explains how it made ChatGPT's data storage 12 times faster to read
Why it matters OpenAI says ChatGPT reads got 12× faster on Azure Cosmos DB with global caching, a design you can copy for high-traffic products.
Impact Caching and multi-tenant storage design offers a template for faster high-traffic reads.
What happened OpenAI describes how it rebuilt Habitat, its online storage system, using multi-tenancy, global caching and Azure Cosmos DB. It says reads for ChatGPT are now 12 times faster.
- 22 OctNewOpenAI explains the safety rules its distribution partners must follow when offering its AI
Impact Resellers and partners offering OpenAI must meet consistent safeguards under its Usage Policies.
What happened OpenAI published a post on how it works with distribution partners to apply the same safeguards across them. It also clarifies what its Usage Policies expect of those partners.
- 21 OctNew
Deployment
A restaurant franchisee used ChatGPT Work to cut price-update time by 95%Why it matters Multi-location operators can test ChatGPT Work on repricing and cash reconciliation, two routine back-office jobs, where one franchisee reports 94-95% less time.
Impact Multi-location operators can automate repricing and cash reconciliation, freeing staff time and cash.
What happened An independent Schlotzsky's and Cinnabon franchisee says ChatGPT Work shortened menu repricing by 95% and cash reconciliation by 94%. The franchisee also set up a process that could release $100K in cash.