AI Against Humanity
AI Against Humanity Logo - A site about AI ethics

AI AGAINST HUMANITY.

Artificial Intelligence is never neutral, because human intention never is. A collection of AI applications that harm, manipulate and threaten humanity. Whether by design, negligence, or indifference. Learn More

Recent Posts

View all posts
Privacy

Concerns Over OpenAI's Codex Background Features

OpenAI's latest update to its Codex desktop app introduces significant new features, including the ability to perform tasks in the background without disrupting user activity. This functionality allows Codex to interact with various applications on a user's computer autonomously, which raises concerns about user privacy and control over their devices. The app can now schedule tasks for future execution and includes an in-app browser for real-time feedback on web development projects. While these advancements aim to enhance productivity, they also blur the lines of user agency and raise ethical questions about AI's role in personal computing. The potential for Codex...

Read article Explore now
Safety

Physical Intelligence, a hot robotics startup, says its new robot brain can figure out tasks it was never taught

Physical Intelligence, a San Francisco-based robotics startup, has introduced its new model, π0.7, which showcases the ability to perform tasks it was not explicitly trained for, a phenomenon known as compositional generalization. This model enables robots to synthesize learned skills from various contexts, allowing them to tackle unfamiliar tasks, and suggests that robotic AI may be on the brink of exponential growth, akin to advancements in large language models. However, researchers caution that π0.7 still has limitations, particularly in executing complex tasks autonomously without detailed human guidance. Despite its impressive capabilities, the lack of standardized benchmarks for robotics raises questions...

Read article Explore now
Ethics

OpenAI takes aim at Anthropic with beefed-up Codex that gives it more power over your desktop

The competition between OpenAI and Anthropic has intensified as both companies vie for dominance in the AI-coding tool market. OpenAI's recent updates to its Codex tool enhance its capabilities, allowing it to operate autonomously in the background on users' desktops. This 'agentic' functionality enables Codex to manage tasks such as frontend changes and app testing, making it a more integral part of corporate workflows. New features, including an in-app browser and a memory function that recalls previous sessions, further improve its utility. However, these advancements raise significant concerns about user autonomy and privacy, as powerful AI tools may operate without...

Read article Explore now
Security

North Korean IT Fraud Threatens U.S. Security

Two U.S. citizens, Kejia Wang and Zhenxing Wang, were sentenced to prison for facilitating a scheme that allowed North Korean IT workers to infiltrate American companies under false identities. The operation involved creating 'laptop farms' that enabled North Koreans to connect to U.S. corporations, leading to the theft of approximately $5 million and sensitive data, including trade secrets from a California-based AI company. The Justice Department emphasized that this fraudulent activity not only harmed U.S. companies financially but also posed a significant risk to national security by embedding North Korean operatives within American computer systems. The scheme highlights the vulnerabilities...

Read article Explore now