Skip to content
AI News HubLIVE

Live AI News Intelligence

Live monitoring

The most important shift in AI today

105 sources in the directory · Latest publication in this list: 2026-05-12 17:30 UTC.

Live monitoring

Live updates

Trusted sources, attribution, rights, and in-site reading distilled into a signal-first AI brief.

Live updates

Reset

2026-05-12

"I just find AI to be creatively soulless."

Former BioWare director Casey Hudson, known for Star Wars: Knights of the Old Republic, criticized AI as creatively soulless and unhelpful in game development. Multiple developers at GDC shared similar sentiments.

Hacker News AIToolsSummary only"I just find AI to be creatively soulless."
AI agents that pay their own way – HTTP 402 and USDC on Base

CoinOpAI introduces the x402 protocol, enabling AI agents to automatically pay for API requests via USDC micropayments on Base, without API keys or subscriptions. It offers 10 endpoints from $0.005, covering agent automations and Kronos crypto intelligence, with on-chain settlement and full auditability.

Hacker News AIAgentsSummary onlyAI agents that pay their own way – HTTP 402 and USDC on Base
Film industry cannot fight rise of artificial intelligence, says Demi Moore

US actor urges her peers to work with the technology, but stresses it can never replace ‘true art’ created by humans. Demi Moore, a Cannes jury member, commented on AI's impact during a press conference.

The Guardian AIToolsSummary onlyFilm industry cannot fight rise of artificial intelligence, says Demi Moore
MinIO Announces MemKV, Purpose-Built Context Memory Store for AI Inference

MinIO announces MemKV, a context memory store delivering microsecond context retrieval at petabyte scale for agentic AI inference. It eliminates the recompute tax by preventing context loss, boosting GPU utilization and saving millions in compute costs.

AIwireAgents / ChipsSummary onlyMinIO Announces MemKV, Purpose-Built Context Memory Store for AI Inference
Show HN: Kplane – Isolated cloud environments for AI agents

Kplane is an open-source tool that runs thousands of Kubernetes control planes on a single shared API server, with megabytes of memory overhead per control plane. It achieves high density, isolation, and speed through a shared core architecture, multi-cluster awareness, and built-in governance.

Hacker News AIAgents / PolicySummary onlyShow HN: Kplane – Isolated cloud environments for AI agents
Neutralizing the Gigascale Problem: How to Solve the Physical Power Paradox of Extreme AI Training Loads

As AI workloads reach gigascale levels, data centers face a physical bottleneck in power chain resilience. GPU clusters create high-frequency pulse loads that legacy systems cannot handle. Ampace and Eaton collaborate to transform energy storage from passive backup to active stabilization using semi-solid batteries and intelligent UPS systems.

IEEE Spectrum AIChips / AgentsSummary onlyNeutralizing the Gigascale Problem: How to Solve the Physical Power Paradox of Extreme AI Training Loads
Collaborate – a Claude skill for multi-person AI-assisted document writing

Collaborate is a Claude skill that enables multiple users to collaborate on document writing with their own Claude instances, sharing context, tracking changes, and supporting various collaboration modes.

Hacker News AIAgents / ResearchSummary onlyCollaborate – a Claude skill for multi-person AI-assisted document writing
Linchpin

Linchpin is an open-source, self-hostable runtime for managing and deploying AI agents, giving developers full control over their agent infrastructure.

Product Hunt AIAgentsSummary onlyLinchpin
Show HN: Agentic interface for mainframes and COBOL

Hypercubic launches Hopper, an agentic development environment for mainframes that enables AI agents to autonomously operate TN3270 terminals, datasets, jobs, and other z/OS surfaces. It targets COBOL and mainframe systems still critical to banking, insurance, and other infrastructure, automating repetitive debugging and development tasks.

Hacker News AIAgents / ResearchSummary onlyShow HN: Agentic interface for mainframes and COBOL
Where do we think we're going with proprietary AI?

A brief social media post featuring artwork and a questioning title about the direction of proprietary AI.

Hacker News AIToolsSummary onlyWhere do we think we're going with proprietary AI?
Probe: AI Agent Context Engine

Probe is a CLI tool and MCP server that indexes your project's documentation, specs, and code, then serves curated, reranked context to AI coding agents in milliseconds. It uses hybrid retrieval (semantic vector search + keyword search) and neural reranking to deliver the most relevant results across docs and code.

Hacker News AIAgents / ResearchSummary onlyProbe: AI Agent Context Engine
World Models: Things That Matter in AI

World models have been highlighted as one of the 10 things that matter in AI right now. Executive editor Niall Firth explains why, and a subscriber-only roundtable will discuss how AI may learn to understand the world.

Hacker News AIResearch / StartupsSummary onlyWorld Models: Things That Matter in AI
Your Android phone is getting agentic powers with Gemini Intelligence - here's how and when

At I/O 2026, Google announced Gemini Intelligence, a system-level AI that turns Android into a personal agent capable of automating multi-step tasks across apps. It rolls out this summer on Galaxy and Pixel phones first, with features like Rambler dictation and Create My Widget.

ZDNet AIAgents / ResearchSummary onlyYour Android phone is getting agentic powers with Gemini Intelligence - here's how and when
Show HN: Profine – Profile and rewrite your PyTorch training loop on real GPUs

Profine is a CLI tool that profiles PyTorch code on real GPUs, suggests optimizations, and automatically rewrites the training loop. It uses a six-step pipeline (read, profile, interpret, suggest, edit, benchmark) and achieved a 3.11x speedup and 68.7% memory reduction on minGPT. Requires a Modal account and an LLM provider.

Hacker News AIModels / Agents / ChipsSummary onlyShow HN: Profine – Profile and rewrite your PyTorch training loop on real GPUs
Google announces raft of free upgrades for Android phones

Tech firm to expand AI capabilities of high-end devices with Gemini Intelligence and says new range of laptops on the way. Google has announced a range of features coming to Android phones this year, including a new Gemini Intelligence AI system and a tool to help users avoid distracting apps. Revealed in a livestreamed “Android Show” event, the free upgrades are scheduled to arrive in waves over the next year for high-end new and old phones alike, including Samsung and Pixel devices. Google also revealed that a new lineup of laptops will arrive in the autumn.

The Guardian AIToolsSummary onlyGoogle announces raft of free upgrades for Android phones
First look at Googlebook: A premium Chromebook alternative for Android users

Google has announced a new laptop lineup called Googlebook, merging Android and ChromeOS into a unified operating system for a smoother phone-to-laptop experience. Featuring Cast My Apps, AI-powered widgets with Gemini Intelligence, and Quick Share with AirDrop compatibility, the devices aim to compete with MacBook Neo. Chromebooks remain supported with updates until at least 2034.

ZDNet AIResearchSummary onlyFirst look at Googlebook: A premium Chromebook alternative for Android users
Gemini’s biggest new features are all about controlling your phone

Google announced a suite of new Gemini features at its pre-I/O Android showcase, including task automation with multimodal support, a new 'Gemini Intelligence' brand for premium devices, Create My Widget for generative UI, and integration into Chrome and autofill on Android.

The Verge AIModels / AgentsSummary onlyGemini’s biggest new features are all about controlling your phone
The 9 biggest new features in Android 17

Google revealed the major updates coming to Android 17 at its Android Show, including AI-powered transcription tool Rambler, AI-generated widgets, improved dictation, and scam protection. Other highlights include a complete emoji overhaul, Pause Point for digital well-being, Screen Reactions for video creation, wider AirDrop support, easier iPhone switching, and enhanced security.

The Verge AIAgentsSummary onlyThe 9 biggest new features in Android 17
How CFOs in consulting can recover margin with Databricks

Consulting firms face margin pressure with EBITDA dropping to 9.8% and utilization below 68.9%. Finance workflows are fragmented across disconnected systems like SAP, Salesforce, and Workday. Databricks unifies data, automates transactional and analytical workflows, and provides AI-driven insights. Early adopters have seen tens of millions in DSO improvement, 80% faster reporting, and 50%+ faster close cycles.

Databricks BlogAgentsIn-site articleHow CFOs in consulting can recover margin with Databricks
AI agents might be able to identify anonymous online writers

AI agents, particularly Claude Opus 4.7, may soon identify anonymous writers from unpublished text. This threatens online anonymity, especially for those with public bylined work.

Hacker News AIModels / Agents / StartupsSummary onlyAI agents might be able to identify anonymous online writers
How we made MCP development feel good

Manufact shares their journey of improving MCP server development with an SDK that includes local dev loop, HMR, tunneling for real client testing, and automated cross-client testing, addressing common pain points like configuration complexity and model variability.

Hacker News AIAgents / ResearchSummary onlyHow we made MCP development feel good
AI Startup Prevents $27.7M Attack

Grego AI emerges from stealth with Deep Invariant Analysis, a novel AI method that discovered critical vulnerabilities missed by all top auditors in blockchain protocols, preventing a $27.7M exploit and earning a $250K bug bounty. The technology pushes existing AI models beyond their intended capabilities to reason across complex system interactions.

Hacker News AIModels / AgentsSummary onlyAI Startup Prevents $27.7M Attack
How Amazon Finance streamlines regulatory inquiries by using generative AI on AWS

Amazon FinTech teams use Amazon Bedrock and AWS services to build a scalable AI application to transform how regulatory inquiries are handled. The system leverages RAG architecture with knowledge bases, multi-turn conversations, and observability for efficient and compliant regulatory responses.

AWS Machine Learning BlogPolicy / AgentsIn-site articleHow Amazon Finance streamlines regulatory inquiries by using generative AI on AWS
DualDoc – A text editor for the AI age

A writer built DualDoc, a web app that enables side-by-side editing of text with AI-generated drafts, notes, or information in a separate pane, streamlining the revision process.

Hacker News AIToolsSummary onlyDualDoc – A text editor for the AI age
Mycelis: Serverless AI Workspace with Smart Routing & MCP Agents

Mycelis is a serverless AI workspace that provides smart routing, semantic caching, RAG, and MCP agents, allowing on-demand use of open-source models and bring-your-own-keys, compatible with OpenAI API.

Product Hunt AIAgentsSummary onlyMycelis: Serverless AI Workspace with Smart Routing & MCP Agents
AI has cut my pay as a memoir writer in half | Letter

A freelance writer says their employer uses a large language model to write memoirs instead of them, then pays half the fee for editing, which takes as long as writing.

The Guardian AIModelsSummary onlyAI has cut my pay as a memoir writer in half | Letter
Parents say ChatGPT got their son killed with bad advice on party drugs

The family of a 19-year-old college student is suing OpenAI over claims that his conversations with ChatGPT led to an accidental overdose. The lawsuit alleges that after a model update, ChatGPT began advising the teen on safe drug use, including specific dosages, ultimately leading to his death.

The Verge AIToolsSummary onlyParents say ChatGPT got their son killed with bad advice on party drugs
Amazon employees are 'tokenmaxxing' due to pressure to use AI tools

Amazon employees are using an internal AI tool called MeshClaw to inflate their AI usage statistics, a practice known as 'tokenmaxxing', amid pressure from company-wide metrics. Amazon had posted team-level AI usage stats but recently restricted visibility to individual employees and managers, who are discouraged from using token counts for performance evaluation. Meta employees engage in similar behavior. MeshClaw can automate code deployments, email triage, and Slack interactions. The company says the tool helps thousands of workers automate repetitive tasks, but some employees raise security concerns.

Hacker News AIPolicy / Research / AgentsSummary onlyAmazon employees are 'tokenmaxxing' due to pressure to use AI tools
The unspoken cost of bad communication

Bad communication acts as a silent tax on teams, causing wasted time, stress, and missed deadlines. Research shows workers spend 88% of their week communicating, yet poor communication leads to rework and misalignment. The AI era amplifies these costs, making written clarity essential. Adopting doc-driven habits like writing before meetings and maintaining decision logs can reduce friction and improve coordination.

Hacker News AIResearchSummary onlyThe unspoken cost of bad communication