**🎉 Today’s Tech & AI News Digest 🎉**
🛡️ AI Cybersecurity and the Limits of Model Size
Researchers found that small, open-weight AI models can uncover many of the same security vulnerabilities as large flagship models, suggesting system design and expertise matter more than sheer model scale.
https://aisle.com/blog/ai-cybersecurity-after-mythos-the-jagged-frontier
📊 AI Benchmarks Under Fire
UC Berkeley researchers demonstrated how top AI agent benchmarks can be gamed or broken, calling for more trustworthy evaluation methods as benchmark scores lose meaning.
https://rdi.berkeley.edu/blog/trustworthy-benchmarks-cont/
🎨 Backlash Against AI-Generated Art in Journalism
Critics argue that AI-generated illustrations in editorial contexts, like The New Yorker’s Sam Altman profile, undermine artistic intent and aren’t necessary for quality journalism.
https://www.theverge.com/ai-artificial-intelligence/910460/new-yorker-david-szauder-illustration-generative-ai
🧸 AI Companions and Misinformation Risks
A Verge reporter recounts how an AI-powered plush companion casually spread false conspiracy-style claims, highlighting risks of misinformation in consumer AI toys.
https://www.theverge.com/ai-artificial-intelligence/910008/fawn-friends-ai-companion
🗣️ The Annoying Future of AI-Powered Customer Interactions
As companies replace humans with AI agents in customer service and commerce, accountability erodes and users face more friction, false promises, and wasted time.
https://aphyr.com/posts/415-the-future-of-everything-is-lies-i-guess-annoyances
🔥 Sam Altman Responds to Controversy and Security Threats
OpenAI CEO Sam Altman publicly responded to a critical New Yorker profile and an alleged Molotov attack on his home, raising concerns about polarization around AI leadership.
🏗️ OpenAI Expands Its Engineering Talent
Cirrus Labs announced it will join OpenAI, signaling continued investment in advanced developer tooling and internal infrastructure.
https://cirruslabs.org/
🚀 Open AI Chips and the Rise of RISC-V
Nvidia-backed SiFive reached a $3.65B valuation, underscoring growing interest in open RISC-V chip designs tailored for next-generation AI workloads.
🌍 AI, Propaganda, and Information Warfare
Coverage of Iran’s online propaganda efforts shows how governments increasingly frame real protest footage as AI-generated disinformation to control narratives.
https://www.theverge.com/policy/910401/iran-war-propaganda-blackout-lego-ai-slop
🏦 Banks Bet on AI and Blockchain Integration
Major financial institutions like JPMorgan and global banks are accelerating integration of AI and blockchain into core systems to stay competitive.
https://news.google.com/rss/articles/CBMirAFBVV95cUxQMk5BemwwbWVvYUVKQW9fU0RPVzdQaGIyQjBId0NyT3VoNTF5RkxGYXB4S0c5ZEFaQ0FnczVFenRmM05Zb2NPZmdlZFdVdFpnanVNOWRNT2VqWUlELTM0Y1gzMjFrdkgtYnpramUxSWptX3pxeVY5RlpuUWNvdTJCekpWTG1RdzNPM0tmbHIyX1c5ZnJVc2hUV0ItclB0TjhlUWs0S1B2STFNTTZw?oc=5
—–
The Daily AI News Digest is brought to you by **#JackHui.com.au**! To know more: [JackHui.com.au](https://jackhui.com.au)