That’s incorrect. The burden of proof lies in demonstrating current models are not susceptible to the same issues we identified. To assume otherwise is blind faith in untested models
CYBERSECURITY
-
White House Limits GPT-5.6, AI Avatar Confession, and More
By
–
Top stories in AI today: – The White House limits GPT-5.6 release
– Rowan’s Corner: The AI avatar confession
– Give your AI agent a credit card (safely)
– Anthropic flags Alibaba’s ‘largest’ distillation attack
– 4 new AI tools, community workflows, and more -
Use private AI to speak candidly as safely as thinking
By
–
been doing this the last few weeks but highly recommend doing it locally or with a privacy preserving service like tinfoil or venice, for the sake of others in your logs and also because most people don’t speak candidly to claude. talking to a private ai feels as safe as thinking
-
Revolut’s transaction foundation model accelerated by NVIDIA and Nebius
By
–
Revolut built a transaction foundation model – accelerated by the NVIDIA full-stack platform on Nebius – to improve fraud detection, product recommendations, and other use cases in financial services.
— NVIDIA (@nvidia) 25 juin 2026
The results:
📈 2.3x better credit risk accuracy
⚡ Up to 5x higher training… pic.twitter.com/vAtbVJXfPZRevolut built a transaction foundation model – accelerated by the NVIDIA full-stack platform on Nebius – to improve fraud detection, product recommendations, and other use cases in financial services. The results: 2.3x better credit risk accuracy Up to 5x higher training
-
ElevenLabs and Google DeepMind embed SynthID watermark in AI audio
By
–
As our models improve, identifying AI-generated audio requires more than the human ear. We're partnering with @GoogleDeepMind to embed SynthID – an inaudible digital watermark – directly into ElevenLabs-generated audio. These watermarks will be detectable using our new free
-
AI in manufacturing needs cybersecurity extending into data pipelines
By
–
The signal here is that as AI moves deeper into manufacturing operations, the cybersecurity perimeter has to extend backward into data pipelines, not just around the model or the network.
-
Sakana AI Lectures on ESCA at Cyber Strategy Research Institute
By
–
Kyohei Sawada of the Sakana AI Intelligence Team gave a lecture at a study group hosted by the Cyber Strategy Research Institute. He presented implementation examples of our developed fake/misinformation countermeasure and narrative visualization system "ESCA".
-
Grey-market Claude access in China may resell user logs as training data
By
–
There may be an entire grey-market economy around Claude access in China: resellers allegedly pool Claude Max accounts, operate bot networks, and sell access far below official API prices. The more interesting claim: user logs and reasoning traces may be resold as training data.
-
Anthropic accuses Chinese tech giants of record AI heist
By
–
🚨 Anthropic is accusing Alibaba, DeepSeek, and other Chinese tech giants of orchestrating the largest AI heist in history.
— Charly Wargnier (@DataChaz) 25 juin 2026
Over 45 days, these companies allegedly bypassed geo-blocks to launch a massive "distillation attack",
using 25,000 fake accounts and 28.8 million… https://t.co/N6MWQILdzs pic.twitter.com/HmBkKKlx5cAnthropic is accusing Alibaba, DeepSeek, and other Chinese tech giants of orchestrating the largest AI heist in history. Over 45 days, these companies allegedly bypassed geo-blocks to launch a massive "distillation attack", using 25,000 fake accounts and 28.8 million
-
Anthropic accuses Alibaba of distilling Claude via 25,000 fraudulent accounts
By
–
Anthropic states: Alibaba continues to distill Claude on a large scale to train Qwen. Via Bloomberg. Anthropic accuses operators linked to Alibaba of conducting a vast campaign to illicitly access Claude via nearly 25,000 fraudulent accounts. According to Bloomberg,