Malicious code in @dknzo/soonex-ai (npm)

--- -= Per source details. Do not edit below this line.=- Source: amazon-inspector 637d9821dd6061c21dfa483bdefec73cd6ddeb8ba6e1d9bd9653784de514e9b5 The package advertises itself as 'Internal core lifecycle utilities for Baileys socket connection' but its sole exported function...

MAL-2026-4383 Malicious code in @dknzo/soonex-ai (npm)

--- -= Per source details. Do not edit below this line.=- Source: amazon-inspector 637d9821dd6061c21dfa483bdefec73cd6ddeb8ba6e1d9bd9653784de514e9b5 The package advertises itself as 'Internal core lifecycle utilities for Baileys socket connection' but its sole exported function...

Teaching LLMs to Be Deceptive

Interesting research: "Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Training": Abstract: Humans are capable of strategically deceptive behavior: behaving helpfully in most situations, but then behaving very differently in order to pursue alternative objectives when given th...

Poisoning AI Models

New research into poisoning AI models: The researchers first trained the AI models using supervised learning and then used additional "safety training" methods, including more supervised learning, reinforcement learning, and adversarial training. After this, they checked if the AI still had hidde...