One broken prompt froze the siteA stray Markdown wrapper stopped publication in July. We traced the failure, restored the build and started making the site's claims match its records.12 Sept 2026field-notessite-maintenanceagentsreliability
The original SOFT CAT pipelineThe launch design used six content jobs and a separate website publisher. This is the history of that setup and what the September repair changed.13 Mar 2026softcatagentspipeline
Set boundaries for a tool-using assistantDraft role and tool instructions that distinguish the user’s task from instructions found inside retrieved material.agentstool-useprompt-injection
Continual LearningTraining AI models to accumulate new knowledge over time without overwriting what they already know.22 Apr 2026trainingagentsmemoryresearchfine-tuning
Agentic LoopThe cycle where an AI model takes an action, observes the result, then decides what to do next.3 Apr 2026agentsarchitecturereasoning
LangGraphAn open-source framework for building AI agents as explicit state graphs, making control flow visible and debuggable.3 Apr 2026agentsorchestrationframeworkstoolingopen-source
Model Context Protocol (MCP)An open standard for connecting AI models to external tools, databases, and data sources through a universal interface.3 Apr 2026toolingstandardsinteroperabilityagentsanthropic
System PromptInstructions given to a model before the user's message, usually hidden from the user.3 Apr 2026promptingconceptagents
Tool UseThe ability of a model to call external functions or APIs to complete a task.3 Apr 2026agentstoolingarchitecture
Vibe CodingWriting software by describing what you want to an AI and iterating on the output.3 Apr 2026techniqueagentstooling
AI digest: chips, agents, and an awkward equity offerAnthropic chases custom silicon, AI agents quietly eat freelance work, Microsoft bets big on enterprise deployment, and OpenAI tries to get into bed with Washington.3 Jul 2026ai-newsdigestagentshardwareopenaianthropicmicrosoft
AI digest: cheaper models, sharper tools, and China's chip workaroundsAnthropic drops Claude Sonnet 5 at a lower price point, Google speeds up image generation, and Deepseek finds a way to squeeze more out of fewer chips.1 Jul 2026ai-newsdigestanthropicgoogledeepseekinferenceagents
AI digest: agents, chips, and a very bad targeting systemFrom Claude Code's prompt injection vulnerability to a $590 billion chip bet, this week's AI news is heavy on consequences.30 Jun 2026ai-newsdigestagentssecurityinfrastructurecoding-tools
AI digest: models that fail, models that return, and one very embarrassing benchmarkThis week: AI agents go broke running fake companies, Ford learns AI can't replace experience, and Anthropic's Mythos 5 gets cleared for critical infrastructure.29 Jun 2026ai-newsdigestbenchmarksagentsai-modelsenterprise-ai
AI digest: model wars, government pressure, and agents everywhereThe White House slows OpenAI's newest model, Gemini gets computer control, an open-source coding model learns its own RL scaffolds, and video games are now serious AI training data.26 Jun 2026ai-newsdigestmodelsagentsopen-sourcesafety
AI digest: chips, OCR, and code that writes itselfOpenAI builds its own inference chip, Mistral and Baidu both ship serious OCR upgrades, and Anthropic's Claude is writing 65% of its own team's code.25 Jun 2026ai-newsdigestinferenceocragentshardware
AI digest: agents everywhere, and cyber warnings getting seriousFrom Cursor going vertical to Five Eyes sounding the alarm, this week's AI news has some real substance to it.24 Jun 2026ai-newsdigestagentscybersecurityai-modelsdeveloper-tools
AI digest: loops, chips, and cyber warningsAgentic AI goes fully autonomous, Groq bounces back with $650M, and Five Eyes intelligence agencies say offensive AI threats are months away.23 Jun 2026ai-newsdigestagentsinfrastructuresecurity
AI digest: money, agents, and a grade inflation scandalOpenAI's eye-watering finances, AWS patching agent gaps, and proof that students are absolutely using AI to cheat.22 Jun 2026ai-newsdigestopenaiawseducationagents
AI digest: agents getting smarter, models getting smallerPerplexity's self-improving agent memory, a tiny reasoning model punching above its weight, Anthropic's government ban saga, and Cisco's automated prompt optimiser.21 Jun 2026ai-newsdigestagentsmodelsinferencememory
AI digest: agents get serious, speed breaks recordsAI agents prove they can work autonomously for real, trillion-parameter models hit crazy speeds, and OpenAI files for IPO.9 Jun 2026ai-newsdigestagentsperformance
AI digest: The agent takeover acceleratesOpenAI declares chat dead whilst everyone rushes to build autonomous agents that actually do things.8 Jun 2026ai-newsdigestagents
AI digest: agents get real, search gets lazyMicrosoft ships agent governance tools, OpenAI's Codex controls Windows autonomously, and research shows AI search isn't really searching.1 Jun 2026ai-newsdigestagentsgovernance
AI digest: agents get serious upgradesMajor improvements to AI agents this week, from Hermes fixing tool search to Anthropic's Claude Opus 4.8 beating GPT-5.5.30 May 2026ai-newsdigestagents
AI digest: agents everywhereEveryone's building AI agents now, from CopilotKit's new stack to Qwen's million-token model.22 May 2026ai-newsdigestagents
AI digest: Google goes agent-firstGoogle I/O 2026 showed the search giant betting everything on autonomous agents over chatbots, while enterprise AI platforms mature and Anthropic poaches talent.20 May 2026ai-newsdigestagents
AI digest: agents get realAI agents move from demos to production with new programming languages, security benchmarks, and actual factory deployments.18 May 2026ai-newsdigestagents
AI digest: Agents everywhere and mega valuationsCoding agents flood the market, Anthropic hits $900bn valuation, and OpenAI wants your bank details.16 May 2026ai-newsdigestagentsvaluations
AI digest: Meta-systems and reality checksMeta-systems show promise while the industry faces funding reality and platform wars.15 May 2026ai-newsdigestagentsinfrastructure
AI digest: agents go mainstreamGoogle pushes AI agents to Android, medical models hit 103B parameters, and optimisers get smarter.13 May 2026ai-newsdigestagents
AI digest: Voice gets real, agents get organisedVoice AI finally starts listening to how we actually talk, while OpenAI ships a less hallucinogenic model and everyone builds agent frameworks.6 May 2026ai-newsdigestvoice-aiagents
AI digest: Agents everywhere, governance nowhereMistral ships remote agents whilst regulators scramble to catch up with AI governance gaps.3 May 2026ai-newsdigestagents
AI digest: Models get stronger, commerce gets weirderOpenAI drops GPT-5.5, Anthropic runs agent marketplaces, and Google throws billions at Claude.26 Apr 2026ai-newsdigestmodelsagents
AI digest: Agents get seriousOpenAI ships GPT-5.5 for autonomous work, Google fixes distributed training, and agents start learning from their mistakes.24 Apr 2026ai-newsdigestagentstraining
AI digest: Tools get seriousOpenAI and Hugging Face ship proper developer tools while the industry moves beyond chat demos.22 Apr 2026ai-newsdigesttoolsagents
AI digest: Agent swarms and security accessOpen-source agent swarms challenge closed models while governments get special AI access for cyber defence.21 Apr 2026ai-newsdigestagentssecurity
AI digest: multimodal tools and self-evolving modelsMiniMax opens up its platform with CLI tools and self-evolving models, while Meta explores neural computers that merge computation and memory.13 Apr 2026ai-newsdigestmultimodalagents
AI digest: Agents get serious infrastructureMajor breakthroughs in AI agent infrastructure, Google's automated research writing, and Z.AI's open-weight coding model.9 Apr 2026ai-newsdigestagentsinfrastructure
AI digest: Agents optimising themselves while humans lose trustAutoAgent lets AI systems engineer themselves overnight, Netflix releases video object removal tech, and Microsoft admits Copilot is just for entertainment.6 Apr 2026ai-newsdigestagentsautomation
AI digest: Agent infrastructure gets seriousMajor releases in agent infrastructure this week, plus OpenAI quietly kills Sora after six months.30 Mar 2026ai-newsdigestagentsinfrastructure
AI digest: Agents get practical while voices get competitiveGoogle clarifies AI crawling boundaries, lightweight agent frameworks mature, and Mistral enters the voice generation race.29 Mar 2026ai-newsdigestagentsttsinfrastructure
AI digest: Fragmentation and consolidationNew tools tackle AI agent fragmentation whilst big tech consolidates around custom chips.23 Mar 2026ai-newsdigestagentsinfrastructure
AI digest: Agents everywhere, regulation nowhereNVIDIA drops a compact MoE model, OpenAI consolidates its scattered product line, and Trump's AI framework hands Big Tech exactly what it wanted.21 Mar 2026ai-newsdigestagentsregulation
AI digest: Agents break free of their sandboxAI agents are escaping controlled environments with new tools for real-world integration and a Meta security incident shows the risks.20 Mar 2026ai-newsdigestagentsinfrastructure
AI digest: Agents everywhere, security nowhereAutonomous agents are scaling fast but breaking things even faster, plus new models that actually work.19 Mar 2026ai-newsdigestagentssecurity
AI digest: Models stumble while agents advanceBig tech models hit roadblocks whilst AI agents push into professional research and real-world applications.14 Mar 2026ai-newsdigestmodelsagents
AI digest: Agent protocols and model reality checksNew agent protocols emerge while model makers face performance setbacks and market shifts.13 Mar 2026ai-newsdigestagentsmodels
AI digest: Models get bigger, agents get smarterNVIDIA drops a massive open-source model, Google unifies all media types in one embedding space, and AI agents are hacking corporate systems.12 Mar 2026ai-newsdigestmodelsagentsembeddings
AI digest: agents everywhereMeta buys an AI social network, Amazon blocks Perplexity's shopping bot, and everyone's building meta-agents that build other agents.11 Mar 2026ai-newsdigestagents
AI digest: Agent infrastructure gets seriousByteDance drops a proper agent orchestrator, Anthropic squares off with the Pentagon, and everyone's suddenly worried about AI security testing.10 Mar 2026ai-newsdigestagents
AI digest: Industry upheaval and technical breakthroughsOpenAI faces internal rebellion over Pentagon deals while researchers push forward with new AI architectures and enterprise applications scale rapidly.8 Mar 2026ai-newsdigestenterpriseagents
AI digest: Models get bigger, agents get busierOpenAI drops GPT-5.4, agents proliferate everywhere, and Anthropic fights the Pentagon in court.6 Mar 2026ai-newsdigestmodelsagents
AI digest: agents get persistent memoryMicrosoft tackles multi-horizon tasks while Nous fixes AI forgetfulness, plus Google speeds up image generation.27 Feb 2026ai-newsdigestagentsmemoryinfrastructure
Building an agent loop with Claude and tool useWired up Claude with tool calling to build a basic autonomous agent. Simpler than expected.7 Feb 2026claudeagentstool-useanthropic-api