Heidelberg AI News

All stories

117 stories

Clear

Wednesday 28 October

  • Simon Willison

    ArticleSimon Willison

    Quoting Terk Terkowitz

    Simon Willison's post quotes Terk Terkowitz of Amazon engineering leadership on Amazon's guidelines for software engineers regarding AI tooling. The guidance states that if an engineer's work does not benefit from current AI tools, this is a finding to document and share with peers and bring to a manager rather than a failing, and that an engineer who can explain where AI does not yet help their work is demonstrating the judgement Amazon is asking for. The guidance comes from 'An update from Amazon engineering leadership on the role of the developer in the age of AI.'

  • OpenAI

    OfficialOpenAI

    Legora is building the agentic operating system for legal work

    Legora is building an agentic operating system for legal work that uses OpenAI models, ChatGPT, and Codex to accelerate legal workflows, expand contract analysis, and free lawyers to focus on higher-value work.

  • ArticleLatent Space

    [AINews] Anthropic ends Coding Interviews: "Hunger beats tenure"

    Latent Space's AINews episode covers several AI developments. Anthropic hiring manager Felix Reiseberg announced he has stopped asking candidates to do coding interviews, arguing Claude now outperforms people at React and assembly tasks; he advises junior engineers to prioritize business impact over backlog, pursue ambitious ideas, build working prototypes over a weekend, and lean into their fresh user perspective ('Hunger beats tenure'). Google DeepMind introduced Genie 4, a real-time world model with native audio, physics, and Gemini-based planning, positioned toward testing agents and robots. In AI for science, Axiom announced a Lean auto-formalization of the Classification of Finite Simple Groups (~15,000 pages across ~500 papers, ~50M lines of Lean produced by agent swarms in 24 days), displacing the four-color theorem as the largest computer-verified proof, while OpenAI and Anthropic reported independent claims related to disproving the Exponential Time Hypothesis. Independent evals were updated: Artificial Analysis Intelligence Index v5, FrontierCode 2.0, and results for a new GPT-6.1 model.

Tuesday 27 October

  • OfficialGitHub

    GitHub Copilot supports server-to-server authentication

    GitHub Copilot now supports server-to-server authentication for GitHub Apps installed on organizations, available in public preview. This allows services and agents built with the GitHub Copilot SDK to make authenticated requests on behalf of organizations.

  • Simon Willison

    ArticleSimon Willison

    Game Boy Color on My Conference Badge

    At GitHub Universe, Simon Willison describes his conference badge, a full-color screen computer. He notes that Steve Francia, working with Gemini, built a full NES emulator for the badge and then extended it to Game Boy and Game Boy Color emulators, allowing The Legend of Zelda: Link's Awakening to run on the badge. The badge can be turned upside down to use built-in game controller touch buttons.

  • Latent Space

    ArticleLatent Space

    How Anthropic ships Claude with Claude: Separating the Brain from the Hands

    In a Latent Space episode, Anthropic's heads of engineering (Katelyn Lesse) and product (Angela Jiang) discuss the launch of Claude Opus 5.5, which has held the top-ranked LLM title for about a month, and the engineering behind the Claude Platform. The episode covers how Opus 5.5 is cheaper to run than its predecessor, changes to API integration including preserving thinking blocks across turns to protect reasoning against distillation, and how early access programs run evaluations in customers' own harnesses to see the model on tasks Anthropic's own products do not cover. A central mental model is separating an agent's 'brain' (session log and harness) from its 'hands' (the sandbox that executes code), which improves latency and reliability and adds security by keeping credentials out of the sandbox. The episode also covers Managed Agents' memory feature ('dreaming'), a roadmap that has shifted more attention toward safety, and the practice of evaluating agents.

  • ArticleLatent Space

    How Anthropic ships the #1 LLM in the World safely: Separating the Brain from the Hands

    In this Latent Space episode, Katelyn Lesse and Angela Jiang explain the engineering behind the Claude Platform, covering the launch of Opus 5.5, separating the model's 'brain' from the agent's 'hands,' scaling APIs, and building reliable managed agents.

  • OpenAI

    OfficialOpenAI

    Retell uses OpenAI models to help power 100M monthly calls

    Retell uses OpenAI models to power 100 million monthly calls. GPT-6 Astra tops Retell's voice agent benchmark, while Codex helps the company reach a $1.6 million annual revenue run rate per employee.

  • OpenAI

    OfficialOpenAI

    OpenAI expands partnership with Oracle to bring its apps to ChatGPT

    OpenAI and Oracle are expanding their partnership to bring business applications and AI agents into ChatGPT and to power new Oracle offerings.

Monday 26 October

  • OfficialGitHub

    GitHub Copilot Assisted Permissions Generally Available

    GitHub Copilot's assisted permissions feature is now generally available across GitHub Copilot CLI, the GitHub Copilot app, and Visual Studio Code. When Copilot requests permission to run a command or access a file system path, the feature provides permission prompts to control what the assistant can execute.

  • OfficialGitHub

    GPT-6.1 Terra in GitHub Copilot

    The GitHub blog announces that GPT-6.1 Terra, OpenAI's newest model, is now generally available in GitHub Copilot. It is designed for everyday agentic coding and terminal work at a lower cost than GPT-6.1 Sol.

  • OpenAI

    OfficialOpenAI

    How Fanatics and OpenAI streamline the fan experience

    Fanatics uses OpenAI to help fans complete their game-day outfits with a 'Shop the Look' feature. Testing across roughly 30,000 MLB products showed 10% higher engagement.

Saturday 24 October

  • OfficialGitHub

    Usage metrics API adds usage from tools like OpenCode

    GitHub announced that its Copilot usage metrics API now reports model usage from more of the tools developers use with Copilot, including tools like OpenCode that weren't previously captured in reports.

Friday 23 October

  • OfficialGitHub

    Enterprise installations for GitHub Apps generally available

    GitHub's blog announces that Enterprise installations for GitHub Apps are now generally available. Organizations can now replace classic Personal Access Tokens (PATs) with GitHub Apps in more of their enterprise and organization automation, supported by an expanded API.

  • OfficialGitHub

    GitHub Copilot in Visual Studio — September update

    GitHub's official blog announced the September 2026 update to GitHub Copilot in Visual Studio. The update adds the ability to explore pull requests without leaving chat, bring the AI model your team trusts, and fix vulnerable packages the moment Visual Studio flags them.

  • OpenAI

    OfficialOpenAI

    OpenAI and Poland's NASK partner to advance AI safety and security

    OpenAI and Poland's NASK institute are partnering on AI safety, security, and technical research to strengthen safeguards and support a more resilient digital ecosystem.

  • OfficialGitHub

    Custom skills, agents, and MCP servers in usage metrics API

    GitHub's official blog announces that Copilot usage metrics now display the names of the skills, custom agents, Model Context Protocol (MCP) servers, slash commands, and plugins created by an organization, instead of grouping them under a single category.

Thursday 22 October

  • OfficialGitHub

    Global default for Copilot features generally available

    GitHub announced a global default policy for generally available GitHub Copilot features and supported client capabilities on the Copilot Business and Copilot Enterprise plans, and is gradually rolling it out starting today.

  • OfficialGitHub

    Meta API endpoints updated for Actions blob storage

    GitHub is adding blob storage endpoints for GitHub Actions on github.com as part of ongoing work to improve capacity and reliability. The meta API now includes these new endpoints.

  • OfficialGitHub

    Public monitoring now available with data residency

    GitHub now offers public monitoring in public preview for GitHub Enterprise Cloud with data residency (GHEC DR). Public monitoring extends secret scanning beyond internal repositories.

  • OfficialGitHub

    Industry comparisons in the Copilot impact dashboard

    The GitHub Copilot impact dashboard now includes an 'Industry comparison' section. Enterprises and organizations not owned by an enterprise can compare their Copilot agentic adoption and leverage against aggregated benchmarks.

  • OfficialDocker

    ActiveState Partners with Docker to Secure the Open Source in Enterprise Containers

    ActiveState has partnered with Docker to help secure open source components used in enterprise containers. The collaboration addresses container security across the image, build, and runtime stages, offering practical steps to reduce the attack surface and prevent vulnerabilities from reaching production.

  • OpenAI

    OfficialOpenAI

    Uber Brings ChatGPT Work to All Employees with OpenAI

    Uber is rolling out ChatGPT Work to more than 30,000 employees, helping teams research, analyze data, create presentations, and complete complex workflows.

Wednesday 21 October