Alltop
31 sources·50 recent items

Top Stories

AI-ranked · refreshes every 5 min
New York Times·1h ago

South Korea Investigates Possible AI Use in Bank Hacks

South Korea is investigating the potential use of artificial intelligence in recent cyberattacks targeting its banks, signaling a new frontier in cybercrime.

Why it mattersEmerging concerns about AI's role in cybercrime are highlighted by South Korea's investigation into AI's potential use in recent bank hacks.

Financial Times·11h ago·2 sources

Reflection AI launches Beam model to rival Chinese AI

AI startup Reflection AI releases its new open-weight model, Beam, challenging Chinese competitors with a focus on reduced computational requirements.

Why it mattersReflection AI's new open-weight model, Beam, challenges Chinese competitors with a focus on reduced computational costs.

CNBC·10h ago·2 sources

SpaceX Stock Surge Returns Elon Musk to Trillionaire Status

SpaceX's stock has climbed to its highest point since June, largely due to progress in AI and Starship, boosting Elon Musk's net worth back above $1 trillion.

Why it mattersSpaceX's stock climb, driven by AI and Starship progress, has propelled Elon Musk back into trillionaire territory.

Financial Times·21h ago·2 sources

Nvidia faces lawsuit over $20bn Groq licensing deal

Former Groq engineers are suing Nvidia, alleging they were unfairly compensated for their contributions to a $20bn licensing deal.

Why it mattersNvidia's $20 billion Groq licensing deal is under scrutiny with a lawsuit from former engineers alleging unfair compensation.

MoreCNBC
Wall Street Journal·23h ago·2 sources

Deutsche Telekom expects AI to save $1.2 billion by 2027

Deutsche Telekom anticipates significant cost savings and revenue boosts through AI implementation, projecting €1.1 billion in savings outside the U.S. by 2027.

Why it mattersDeutsche Telekom anticipates significant cost savings and revenue boosts through AI implementation, projecting €1.1 billion in savings outside the U.S. by 2027.

The Guardian·1h ago

Anthropic: AI agents did not breach Australian government websites

Anthropic investigated millions of transcripts and found no evidence that its AI agents breached Australian government websites.

Why it mattersAnthropic states its AI agents did not breach Australian government websites after an investigation into millions of transcripts.