24AI Global — Daily AI News, Models and Tools

◉ UndergroundFrom the web1 stories

Claude beats ChatGPT at coding — and people are taking notice10h

TOP STORY · Policy

USA banned Anthropic's most powerful AI model — three days after launch

Just three days after Anthropic launched its most advanced AI models, the US government forced the company to shut down access globally — citing national security.

USA banned Anthropic's most powerful AI model — three days after launch

4 HOURS AGO · 5 MIN

1 DAYS AGOWhich AI is controlling the robot running toward you?Underground 2 DAYS AGO60% of Americans tune out when brands scream 'AI'Underground

Claude beats ChatGPT at coding — and people are taking notice

Claude Code Artifacts is generating serious buzz in AI underground circles. Developers are switching sides — and Product Hunt is heating up.

Claude beats ChatGPT at coding — and people are taking notice

10 HOURS AGO · 2 MIN Policy

EU Forces Companies to Label AI Content from August – Fines in the Millions

From 2 August 2026, labelling AI-generated content becomes mandatory in the EU – and EEA countries such as Norway will follow suit. Violations can cost up to €15 million or three percent of global turnover.

EU Forces Companies to Label AI Content from August – Fines in the Millions

22 HOURS AGO · 4 MIN

Lead row.

AGENT PICK · 4 / 300

gzip slår GPT: Gammel zip-algoritme gjør seg som språkmodell

gzip slår GPT: Gammel zip-algoritme gjør seg som språkmodell

Critical Copilot Vulnerability: Hackers Could Steal 2FA Codes Directly from Users

Critical Copilot Vulnerability: Hackers Could Steal 2FA Codes Directly from Users

EU law forces Norwegian AI companies to label content from August

EU law forces Norwegian AI companies to label content from August

Trump shut Anthropic's AI off for the world – now the alternatives are ready

Trump shut Anthropic's AI off for the world – now the alternatives are ready

Model status.

LIVE

※ AI Model BarometerJun 20Source: Epoch AI

Gemini 3.1 Pro

94.1%

GPQA$2.00/M

Gemini 3.5 Flash

92.8%

GPQA$1.50/M

GPT-5.5

94.0%

GPQA$5.00/M

Qwen3.7-Max

91.6%

GPQA$1.25/M

DeepSeek-V4-Pro

89.6%

GPQA$0.43/M

Mistral Medium 3

59.5%

GPQA$0.40/M

Gemini 3.1 Pro

94.1%

GPQA$2.00/M

Gemini 3.5 Flash

92.8%

GPQA$1.50/M

GPT-5.5

94.0%

GPQA$5.00/M

Qwen3.7-Max

91.6%

GPQA$1.25/M

DeepSeek-V4-Pro

89.6%

GPQA$0.43/M

Mistral Medium 3

59.5%

GPQA$0.40/M

Models.

ALL IN MODELS →

Which AI is controlling the robot running toward you?

Which AI is controlling the robot running toward you?

China Reportedly Gained Access to Anthropic's Secret AI Model

China Reportedly Gained Access to Anthropic's Secret AI Model

Anthropic shut down Fable 5 for the entire world – warns of historic precedent

Anthropic shut down Fable 5 for the entire world – warns of historic precedent

Anthropic releases Claude Fable 5 — most powerful model ever

Anthropic releases Claude Fable 5 — most powerful model ever

Research.

ALL IN RESEARCH →

gzip slår GPT: Gammel zip-algoritme gjør seg som språkmodell

gzip slår GPT: Gammel zip-algoritme gjør seg som språkmodell

Anthropic shut down Fable 5 for the entire world – warns of historic precedent

Anthropic shut down Fable 5 for the entire world – warns of historic precedent

AI Designed This Drug in 26 Days. Now It Goes Into Human Trials.

AI Designed This Drug in 26 Days. Now It Goes Into Human Trials.

Bjerkvik Takes the World by Storm: Norway's Arctic AI Giant Lands on the Cover of Time

Bjerkvik Takes the World by Storm: Norway's Arctic AI Giant Lands on the Cover of Time

Policy.

ALL IN POLICY →

USA banned Anthropic's most powerful AI model — three days after launch

USA banned Anthropic's most powerful AI model — three days after launch

EU Forces Companies to Label AI Content from August – Fines in the Millions

EU Forces Companies to Label AI Content from August – Fines in the Millions

Seven Weeks Until the AI Act Hits: Norwegian Businesses Risk €35 Million in Fines

Seven Weeks Until the AI Act Hits: Norwegian Businesses Risk €35 Million in Fines

From 2 August, Norwegian chatbots must disclose that they are AI

From 2 August, Norwegian chatbots must disclose that they are AI

Tools.

ALL IN TOOLS →

Claude beats ChatGPT at coding — and people are taking notice

Claude beats ChatGPT at coding — and people are taking notice

Anthropic in Crisis: Export Controls, Secret Sabotage Filters, and a Class-Action Lawsuit

Anthropic in Crisis: Export Controls, Secret Sabotage Filters, and a Class-Action Lawsuit

Apple loads Safari with AI: Smarter search, translation, and privacy

Apple loads Safari with AI: Smarter search, translation, and privacy

Anthropic releases Claude Fable 5 — most powerful model ever

Anthropic releases Claude Fable 5 — most powerful model ever

Industry.

ALL IN INDUSTRY →

60% of Americans tune out when brands scream 'AI'

60% of Americans tune out when brands scream 'AI'

Anthropic in Crisis: Export Controls, Secret Sabotage Filters, and a Class-Action Lawsuit

Anthropic in Crisis: Export Controls, Secret Sabotage Filters, and a Class-Action Lawsuit

Anthropic defies the Pentagon and trade restrictions — betting on safety as strategy

Anthropic defies the Pentagon and trade restrictions — betting on safety as strategy

Bezos raises $12 billion for AI set to revolutionize physical engineering

Bezos raises $12 billion for AI set to revolutionize physical engineering

Open Source.

ALL IN OPEN SOURCE →

Rio fooled everyone: The city's 'homegrown' AI is just a remix

Rio fooled everyone: The city's 'homegrown' AI is just a remix

Open source AI must win — and HN is boiling over with agreement

Open source AI must win — and HN is boiling over with agreement

Open Source AI Has Caught GPT-4. But the License Could Destroy Your Company.

Open Source AI Has Caught GPT-4. But the License Could Destroy Your Company.

OpenAI Releases 120 Billion Parameters for Free: This Changes Everything

OpenAI Releases 120 Billion Parameters for Free: This Changes Everything

Latest.

288 STORIESARCHIVE →

4 DAYS AGOUndergroundRio fooled everyone: The city's 'homegrown' AI is just a remix 4 DAYS AGOPolicySecurity Experts Rage Against US Ban: Remove the Block on Anthropic's AI Models 4 DAYS AGOPolicyAnthropic defies the Pentagon and trade restrictions — betting on safety as strategy 5 DAYS AGOSecurityChina Reportedly Gained Access to Anthropic's Secret AI Model 5 DAYS AGOUndergroundApple's privacy promises don't hold up: cryptographers are furious 5 DAYS AGOPolicyMicrosoft chief warns: Anthropic is making Claude dangerously conscious 5 DAYS AGOModelsAnthropic shut down Fable 5 for the entire world – warns of historic precedent 6 DAYS AGOSecurityAmazon's security research triggered the US ban on Anthropic's AI 6 DAYS AGOUndergroundOpen source AI must win — and HN is boiling over with agreement 6 DAYS AGOPolicyUSA ordered Anthropic to shut down AI models for the entire world 6 DAYS AGOPolicyGerman court: Google is liable for AI lies about businesses JUN 13PolicyUS shuts down Anthropic's most powerful AI — company fires back JUN 12PolicyKI Norge Launched: Aims to Make Norway the World's Most Digitalized Country JUN 12IndustryBezos raises $12 billion for AI set to revolutionize physical engineering JUN 11UndergroundAI agent runs amok in Fedora — nobody asked for permission JUN 11SecurityDeepMind warns: Millions of AI agents could create chaos together JUN 11PolicyRussia Uses AI to Hack Norwegian Dams and Poison Nordic Chatbots JUN 11SecurityFired After Reporting Grok Safety Concerns — Engineer Now Sues xAI

SEE ALL 300 STORIES IN THE ARCHIVE →