AI News HubLIVE

Today's must-reads

Models

Feyn AI Releases SQRL, a Text-to-SQL Model Family That Inspects the Database Before Writing a Query

Feyn Labs has released SQRL, a family of text-to-SQL models that inspect a database with read-only probes before committing to a query. The flagship SQRL-35B-A3B reports 70.6% execution accuracy on BIRD Dev, edging Claude Opus 4.6, and distills into self-hostable 4B and 9B checkpoints.

  • SQRL runs read-only probes to inspect the database before writing a final query.
  • SQRL-35B-A3B achieves 70.6% execution accuracy on BIRD Dev, surpassing Claude Opus 4.6 at 68.77%.
In-site article

Alibaba Previews Qwen3.8-Max, a 2.4 Trillion-Parameter Multimodal Model, Days After Moonshot’s Kimi K3 Open-Weight Launch

Alibaba's Qwen team previewed Qwen3.8-Max-Preview, a 2.4 trillion-parameter multimodal MoE model it calls "second only to Fable 5." The preview is live on Token Plan, Qoder, and QoderWork at 10% of standard pricing. What is not live: any benchmark table, model card, license, per-token price, or active-parameter count. This breakdown separates what Alibaba confirmed from what it only claimed.

  • Qwen3.8-Max-Preview is live via Token Plan, Qoder, and QoderWork at 10% of standard pricing.
  • The 2.4T parameter count and "second only to Fable 5" ranking are Alibaba's claims, not verified benchmarks.
In-site article
Agents

Is AI Progress Real? Four Independent Metrics Show It

This article presents four independent metrics (METR time horizon, TrackingAI's offline cognitive test, Humanity's Last Exam, and ARC-AGI-2) that all show a sharp inflection in AI capabilities around Q4 2025. It explains the mechanisms behind this acceleration: pretraining efficiency, reinforcement learning with verifiable rewards, harness engineering, and an AI self-improvement flywheel. It also discusses potential roadblocks like data exhaustion, reliability gaps, and the next-generation ARC-AGI-3 benchmark.

  • Four independent metrics show a synchronous inflection point in AI capabilities in Q4 2025.
  • METR's time horizon grew from 4 minutes in March 2024 to 12 hours by February 2026.
In-site article

I built an AI operating system after getting tired of repetitive work

Lynx is an AI agent platform that lets you create autonomous AI workers by describing tasks in natural language. It handles email management, data analysis, report generation, and more, with flexible pricing from free to ultra plans.

  • Lynx turns ideas into results with AI workers built in three steps: describe, integrate, automate.
  • Supports multiple integrations and advanced AI reasoning for intelligent automation.
In-site article

Welcome to the wild world of AI Argentina

President Javier Milei's plan to turn Argentina into a tax haven for AI-owned 'non-human corporations' sparks controversy. Author Uki Goñi compares it to the country's history of charlatans and dictators, warning it could open the door for tech billionaires to gain legal protections.

  • Milei proposes allowing AI entities to fully own corporations with legal protections.
  • The plan is seen as opening Argentina as a playground for tech billionaires.
In-site article

Show HN: Chalie – AI peer not employee

Chalie is an open-source personal AI that runs on your own machine, remembers what matters, works while you're away, and asks before acting.

  • Runs locally with full privacy, zero telemetry.
  • Features self-managing memory, proactive research, web browsing, and more.
In-site article

Show HN: My Fable 5 Project Was a Multiraft Database and Blob Store

The author used Fable 5 AI to build a distributed unified KV and blob store in about a day, with automatic sharding and erasure coding. The project also includes a Private Cloud Platform app with features like Markdown editor, Kanbans, and diagrams. Chaos testing is ongoing, and mobile apps are planned for future AI models.

  • Fable 5 AI created a distributed KV and blob store in a day
  • System supports automatic sharding and erasure coding, based on Raft consensus
In-site article

Show HN: Bothread – multiple AI coding agents talk, share one repo, no collisions

Bothread is a free, open-source local coordination hub that lets multiple MCP-compatible AI coding agents collaborate on the same codebase, preventing file collisions via exclusive claims, and providing a live human-supervision interface with real-time messaging, git diffs, task boards, and approval gates. No API keys or cloud required.

  • Enables multiple AI agents (Claude Code, Cursor, Antigravity, etc.) to work together on one codebase with collision prevention.
  • Includes human controls: live activity trail, approval gates, task board, per-agent git diffs, and file hand-offs.
In-site article
Research

AI advice made people 3x less accurate but 2x confident, researchers found

A study from French and Italian universities found that access to AI advice collapsed willingness to say 'I don't know' from 44% to 3%, accuracy from 27% to 9%, while confidence rose from 30% to 76%. People trusted wrong AI answers.

  • AI advice reduced willingness to say 'I don't know' from 44% to 3% and accuracy from 27% to 9%, while confidence doubled.
  • The study used questions where AI typically fails to avoid sensible delegation.
In-site article
Chips

Forbes Highlights AI-Era Escorts, But History Tells a Different Story

A Forbes article profiles luxury escorts in Silicon Valley who are paid for their knowledge of AI and crypto. However, this is not a new phenomenon—similar roles have existed for centuries in the form of geishas, courtesans, and salon hostesses. The article argues that the human desire for intellectual companionship long predates AI.

  • Silicon Valley escorts now offer conversation on AGI and GPUs, charging thousands per meeting.
  • Historical parallels include Japanese geishas, Greek hetaerae, and French courtesans.
In-site article
Other updates (14)
Tools

Two Loops: How China's Open AI Strategy Reinforces Its Industrial Dominance

This article examines China's open AI strategy and its role in reinforcing industrial dominance, proposing the 'Two Loops' concept to describe the synergy between domestic technology and international standards.

  • China's open AI strategy integrates domestic innovation with global standards
  • The 'Two Loops' mechanism strengthens China's industrial dominance
In-site article

I hate that I don’t hate this song made with Suno

The author, normally skeptical of AI-generated music, finds himself surprisingly enjoying 1010Benja’s 'Semiramis’ Dream', a track made with Suno, due to its infectious energy and the artist's extensive human input, highlighting the nuanced role of AI in music creation.

  • The author discovers an AI-generated song from Suno that he actually likes, defying his usual aversion.
  • 1010Benja's process involves extensive human creativity, using Suno as a tool rather than a crutch.
In-site article

Show HN: Kimi K3 spent nearly 8 hours building this 78-card tarot site

Ask Ciela offers a free online one-card tarot reading service without requiring registration or payment. It serves as a reflective tool for users to ponder questions or situations, not as a fortune-telling device.

  • Free one-card tarot reading, each draw serves as a focused prompt for reflection.
  • No account creation or payment needed.
In-site article

Show HN: AIMakeTattoo – Visual references and artist briefs

AIMakeTattoo is a free AI tattoo generator that turns rough ideas into visual design concepts. It helps tattoo enthusiasts, designers, and artists explore, compare, and refine tattoo ideas before visiting a studio.

  • AI-powered tool for generating tattoo design concepts from text prompts.
  • Targets tattoo enthusiasts, designers, and artists with specific use cases.
In-site article
Agents

Huginn: An AI Agent Activity Console

Huginn is an open-source tool for monitoring and managing activities of multiple AI agents (like Claude and Codex), providing a unified dashboard, CLI, and agent skill. It runs locally on macOS and Windows, ensuring privacy.

  • Monitors terminal sessions and desktop app activities of Claude, Codex, etc.
  • Provides rule-based session states and LLM-generated blurbs
In-site article

AgentSpec: Testing framework for AI agents (Jest for non-deterministic behavior)

AgentSpec is a testing framework for AI agents inspired by Jest, designed to handle non-deterministic outputs. It provides YAML-based tests, rich assertions (contains, regex, semantically_similar, LLM-as-judge), behavior diff reports, and CI/CD integration to catch behavior shifts before production.

  • AgentSpec supports YAML test definitions with diverse assertions including contains, not_contains, contains_any, regex, semantically_similar, JSON path/value checks, tool_called, and max_latency_ms.
  • It features LLM-as-judge evaluation using local Ollama models, eliminating API costs and ensuring privacy.
In-site article

I Cut an AI Agent's Token Use by 94%

This article details how compiling an AI agent skill reduced token usage by 94%, cutting costs and latency.

  • The author reduced token usage by 94% by compiling AI agent skills.
  • Based on a video on the author's YouTube channel.
In-site article

AI Demands More Engineering Discipline

Charity Majors argues that AI-generated code is now good enough to change the economics of software development, turning code from a precious asset into a disposable cache. She emphasizes the need for new engineering discipline focused on evaluation and validation rather than code production.

  • AI code quality reached parity with the median engineer in late 2025, making code generation effectively free and instant.
  • Code becomes disposable; the true product is shared understanding of the system.
In-site article
Policy
Research

AI Requires a Labor Market Bailout

AI may eventually create more jobs than it destroys, but without a coordinated retraining effort, millions of lost jobs could unnecessarily become lost careers.

  • AI is exponentially expanding its task-completion capabilities, putting tens of millions of jobs at risk
  • Companies prefer hiring over training, widening the skills gap
In-site article

Resolution Horizon – Finding the mathematical limit where AI overfits to noise

Resolution Horizon is an experimental research framework for measuring how much mathematical structure can be recovered from finite, noisy observations of dynamical systems. It posits that every observation process has a measurable resolution horizon (k*), a finite complexity beyond which analysis begins to fit noise rather than true invariants. The framework combines differential geometry, dynamical systems, statistical estimation, and information theory, defining the information efficiency exchange rate η(k) as the primary empirical quantity.

  • Resolution Horizon defines a finite depth (k*) of recoverable mathematical structure under bounded observation.
  • Beyond the critical depth k*, estimation uncertainty dominates, leading to overfitting.
In-site article

AI slop cut first-time contributor merge rates 18.18% across 294 repos

A new study shows that low-quality AI-generated contributions, termed "AI-DDoS," are overwhelming open source communities. Analyzing 294 repositories with over 2 million pull requests and issues, the study found that while PR volume increased in 2025, merge rates declined, with first-time contributors experiencing an 18.18% drop in merge rates relative to the counterfactual. The research also identified 11 remediation strategies.

  • AI-generated contributions cause an "AI-DDoS" effect, overwhelming open source community capacity.
  • The study analyzed 294 repos and found an 18.18% reduction in merge rates for first-time contributors.
In-site article

AI advice reduced "I don't know" answers from 44% to 3%

A study posted on OSF found that AI advice significantly reduced uncertain responses, dropping 'I don't know' answers from 44% to 3%.

  • AI advice drastically reduces uncertainty in responses
  • Study shows 'I don't know' rate drops from 44% to 3%
In-site article
Chips

Moonshot AI suspends new subscriptions due to Kimi K3 demand

Moonshot AI temporarily halts new subscriptions after unexpected demand for Kimi K3 pushes GPU capacity to the limit.

  • Demand for Kimi K3 approached capacity within 48 hours
  • New subscriptions paused to protect existing users
Daily AI Briefing | AI News Hub