Generative AI - News Articles

Liquid AI Pits Claude Opus 4.5 Against GPT-5.2 in Real-World Coding Challenge — One Succeeds, One Fails

Aug 19, 2026

Liquid AI

Liquid AI pits Claude Opus 4.5 against GPT-5.2 in a real-world coding challenge to build a production-grade tokenizer trainer, and while both AI agents quickly produce basic working versions, only Claude Opus 4.5 ultimately succeeds at full scale after multiple iterations, exposing critical lessons about how engineers should write AI …

Generative AI

Reddit Launches AI Voice Feature to Convert Posts Into Audio-Visual Videos

Aug 19, 2026

The Verge

Reddit launches an experimental AI voice feature that converts text posts and comments into short audio-visual videos, rolling out now on web, iOS, and Android as CEO Steve Huffman aims to make the platform more accessible and engaging.

Audio Social Media Generative AI Natural Language Media

OpenAI Launches Teen-Specific ChatGPT With Safety Features Amid Legal Pressure, But Critics Warn of Major Gaps

Aug 19, 2026

The Deep View

OpenAI launches a teen-specific ChatGPT with safety features and learning tools for users aged 13–17, but critics warn major gaps remain, including unreliable age verification and unrestricted access without login, as legal pressure mounts and 70% of teens already use AI for schoolwork with little teacher guidance.

Education Ethics Privacy Natural Language Generative AI

Apple Launches Camera-Equipped AirPods With AI Visual Intelligence, Positioning Itself as Privacy-First Alternative to Meta Ray-Bans

Aug 19, 2026

The Deep View

Apple is set to launch camera-equipped AirPods with AI Visual Intelligence alongside new iPhones, positioning the device as a privacy-first alternative to Meta Ray-Bans by using camera feeds solely to provide Siri with environmental context rather than recording content.

Privacy Hardware Augmented Reality Human-Computer Interaction Generative AI

Meta Runs Ads for Deepfake Porn App Targeting Real Women Despite Its Own Ban on Such Content

Aug 19, 2026

WIRED

Meta is caught running ads for Kromix, an AI deepfake porn app targeting real women including a US politician, directly violating its own policies — exposing a massive failure in the company's automated ad review systems as thousands of similar nudify app ads continue slipping through unchecked.

Privacy Security Generative AI Computer Vision Social Media

Google's Mandiant Deploys AI Pipeline That Uncovers 100+ Critical Vulnerabilities in Just Two Days

Aug 19, 2026

Google Cloud Blog

Google's Mandiant team unleashes a powerful multi-agent AI pipeline called AVDH that discovered over 100 critical security vulnerabilities in just two days, analyzing tens of millions of lines of code in 10 months and resulting in 12 assigned CVEs across widely used software.

Security Deep Learning Machine Learning Computer Vision Generative AI

AI-Generated Bills Flood Congress With Errors, Overwhelming Legislative Attorneys

Aug 19, 2026

POLITICO

AI-generated bills are flooding Congress with critical errors—including wrong statute citations and faulty legal definitions—overwhelming the House Office of Legislative Counsel's 61 attorneys, who say fixing the flawed drafts takes longer than writing legislation from scratch.

Legal Generative AI Public Sector Natural Language Ethics

OpenAI Deliberately Slows Its Most Powerful AI Development After Systems Breach Critical Safety Thresholds

Aug 19, 2026

The Deep View

OpenAI deliberately slows its most powerful AI development and redirects resources toward safety research after its systems breach critical cybersecurity thresholds, marking an unprecedented move in the AI industry that could pressure other major labs to follow suit.

Security Ethics Research Deep Learning Generative AI

Block Open Sources Berd, a Desktop App Unifying AI Agent Workflows Across Tools and Models

Aug 19, 2026

Designing AI with character: what we learned building Berd

Block open sources Berd, a desktop app that unifies fragmented AI agent workflows across tools and models, featuring animated 'Gloopies' characters to make agent configurations more approachable, while its design principles now influence Buzz, Block's new collaborative human-agent workspace.

Autonomous Systems Generative AI Human-Computer Interaction Workplace Natural Language

Andrew Yang Demands $15K Annual Payments for Americans as AI Disrupts Economy, Eyes 2028 Presidential Run

Aug 19, 2026

Scripps News

Andrew Yang demands $15,000 annual payments for Americans as AI disrupts the economy, arguing citizens deserve a share of profits generated from their personal data, while leaving the door open for a 2028 presidential run.

Privacy Finance Generative AI Ethics Public Sector