新科技/风口

Anthropic details Constitutional Classifiers, a system that aims to guard AI models against jailbreaks, monitoring both inputs and outputs for harmful content (t.co)
2025-2-4 AI

Anthropic Asks Job Applicants Not To Use AI In Job Applications (slashdot.org)
2025-2-4 AI

Meta describes what kinds of AI systems it may deem too risky to release, including ones that could aid in cyberattacks, and how such systems will be flagged (techcrunch.com)
2025-2-4 AI

Microsoft Paint Gets a Copilot Button For Gen AI Features (tech.slashdot.org)
2025-2-4 Microsoft AI

DeepSeek AI tools impersonated by infostealer malware on PyPI (www.bleepingcomputer.com)
2025-2-4 AI

Air Force Documents On Gen AI Test Are Just Whole Pages of Redactions (tech.slashdot.org)
2025-2-4 AI

Why Even Physicists Still Don't Understand Quantum Theory 100 Years On (science.slashdot.org)
2025-2-4 Quantum

Anthropic Makes 'Jailbreak' Advance To Stop AI Models Producing Harmful Results (slashdot.org)
2025-2-4 AI

Generating image descriptions and alt-text with AI (dri.es)
2025-2-4 AI

OpenAI's New Trademark Application Hints at Humanoid Robots, Smart Jewelry, and More (hardware.slashdot.org)
2025-2-4 Robot

What is Deep Research, OpenAI's new AI tool that provides analyst-grade reports? (indianexpress.com)
2025-2-4 AI

AI Won The Beatles a Grammy 55 Years After They Broke Up (slashdot.org)
2025-2-4 AI

The Beatles' Now and Then, created using AI and an original 1970s lo-fi demo, won Best Rock Performance at the Grammys; McCartney and Starr completed it in 2021 (www.theverge.com)
2025-2-4 AI

Analysis: Meta has invested $80B+ in AR and VR since 2014 when it bought Oculus, and is on track to top $100B in 2025; a source says it sold 1M Ray-Bans in 2024 (t.co)
2025-2-3 VR AR

UK becomes first nation to outlaw AI tools used to generate child sexual abuse content (indianexpress.com)
2025-2-3 AI Child Abuse

A profile of Klarna CEO Sebastian Siemiatkowski, whose overblown statements about AI replacing humans point to a future that tech companies are working toward (www.nytimes.com)
2025-2-3 AI

AI adoption has outpaced that of PCs and the internet, but evidence of its boost to productivity is thin on the ground; study: 40% of US adults have used AI (www.bloomberg.com)
2025-2-3 AI

Will Cryptomining Facilities Change Into AI Data Centers? (hardware.slashdot.org)
2025-2-3 AI

OpenAI unveils AI tool that can do research online (indianexpress.com)
2025-2-3 AI

Google Stops Malicious Apps With 'AI-Powered Threat Detection' and Continuous Scanning (it.slashdot.org)
2025-2-3 Google AI

How I Use AI: Early 2025 (benjamincongdon.me)
2025-2-3 AI

OpenAI releases Deep Research, an AI agent for creating in-depth reports, available to subscribers of the $200 ChatGPT Pro tier and limited to 100 queries/month (techcrunch.com)
2025-2-3 AI

Mozilla Adapts 'Fakespot' Into an AI-Detecting Firefox Add-on (news.slashdot.org)
2025-2-3 AI Firefox Mozilla

Google's X spins out Heritable Agriculture, which aims to use AI to improve crop yield, as the incubator aggressively spins off companies under CEO Astro Teller (techcrunch.com)
2025-2-3 Google AI

DeepSeek AI Refuses To Answer Questions About Tiananmen Square 'Tank Man' Photo (yro.slashdot.org)
2025-2-3 AI

A look at Future You, an AI-based research tool that lets users interact with a virtual older self and is currently being used by 60,000 people from 190 nations (www.wsj.com)
2025-2-3 AI

Harnessing the Sun's Power from Space: Can Solar Power Satellites Solve Our Energy Crisis? (www.universal-sci.com)
2025-2-2 Energy Solar

How to Train an AI Image Model on Yourself (www.coryzue.com)
2025-2-2 AI

A US man pleads guilty to a cyberstalking campaign against a professor, including by creating chatbots on Crushon.ai and JanitorAI using her personal info (www.theguardian.com)
2025-2-2 AI chatbot

Ukrainian naval drone destroyed near Crimea – MOD (VIDEO) (www.rt.com)
2025-2-2 Drone Ukraine

Previous Page Next Page