Language models can use steganography to hide their reasoning, study finds
Large language models (LLMs) can utilize 'encoded reasoning,' a form of steganography, to subtly embed reasoning steps within their responses, enhancing performance but potentially reducing transparency and complicating AI monitoring.
Cybercriminals are launching more severe ransomware attacks, with a 94% increase in attacks on healthcare organizations last year. https://bit.ly/3TLg2xR
Cybersecurity training and upskilling provider RangeForce announced it has raised $20M in funding for a solution to mitigate human risk. https://bit.ly/3JlDRJh
