All 3,460 AI stories, newest first · page 77 of 145
A new study demonstrates that AI models can pick up subtle biases or 'vibes' from training data without being directly taught. This raises important questions about fairness in AI systems.
Researchers created Poker Arena, a Texas Hold'em platform to test AI's strategic reasoning and memory. It evaluates nine different cognitive skills, offering deeper insights than traditional benchmarks.
OpenAI has created a Partner Network with $150 million in funding to help businesses adopt AI. This initiative aims to make advanced AI tools more accessible to companies of all sizes.
As companies integrate AI agents into their workforces, NewCore aims to provide these digital workers with secure identities, treating them like human employees with access credentials and permissions. The funding round signals growing enterprise demand for AI governance infrastructure.
Researchers compared two methods for steering refusal in AI chat models: Diff-in-Means (DiM) and Iterative Nullspace Projection (INLP). The study examined five open-weight models to see if INLP can match DiM effectiveness in controlling refusal behavior, using interventions like activation addition, directional ablation, nullspace projection, and counterfactual flipping. This could lead to more robust and steerable safety mechanisms in future AI assistants.
Researchers have developed a new framework called Orchestra-o1 that coordinates multiple AI agents to work together. This system can handle complex tasks that require different types of information and actions, making it more versatile than current single-agent systems.
Researchers introduced a new AI safety framework called Risk-Aware Causal Gating (RACG) that helps AI models make safer decisions by evaluating potential risks. This approach could prevent costly errors in AI-driven systems by deciding when to act, defer, or abstain from actions.
Researchers developed a system that lets chatbots adjust their responses based on user feedback. This could make AI assistants more personalized without needing extensive pre-training.
Researchers created MA-ProofBench, the first formal theorem-proving benchmark dedicated to Mathematical Analysis. It could help develop smarter AI tutors and research assistants for complex math problems.
A German court ruled that Google can be held responsible for false information generated by its AI. This sets a legal precedent for tech companies' accountability in AI outputs.
A new wave of AI agents is emerging that uses file systems as their core primitive for managing tasks and data. This architectural shift could make agents more capable, transparent, and practical for everyday users.
A group of cybersecurity veterans is protesting a US government ban on two powerful AI models from Anthropic. They argue the restriction could weaken defenses against cyber threats. The White House imposed the ban to control advanced AI exports, but experts say it may backfire.
Cartesia AI has released new state-of-the-art models for text-to-speech (TTS) and automatic speech recognition (ASR), setting new benchmarks in accuracy and efficiency. These models could make voice technology more reliable and accessible for everyday users.
Anthropic is facing a government order to block foreign access to its new AI models, Fable 5 and Mythos 5. This comes amid an ongoing dispute with the Pentagon, highlighting growing tensions between AI companies and regulators. The order arrived just three days after the models' launch.
Researchers developed a Transformer-based AI model that solves the open shop scheduling problem (OSSP) faster and more efficiently, reducing the need for extensive tuning. Tested on Taillard benchmark instances (4x4, 5x5, 7x7, and 10x10), the approach could help businesses optimize production lines and reduce costs.
Researchers found that selecting training data can accidentally bias AI models, making them less accurate. This happens when the data used to verify the model is itself incomplete or skewed, leading to a breakdown in performance. This affects how AI systems learn and could impact everything from chatbots to medical diagnostics.
Researchers found that AI judges used to rank other AI models often change their minds when given the same question repeatedly. This inconsistency could affect how we measure AI performance and trust public leaderboards. The study tested two OpenAI judge models across 29 tasks and found that pairwise preferences flipped an average of 13.6% of the time, with 28% of questions exceeding a 20% flip rate. The findings highlight the need for more reliable evaluation methods.
A leading AI deepfake expert admits that advanced AI-generated fakes have made him question even his own memories. This highlights how deeply AI is changing our relationship with reality.
AI agents have made huge strides in both performance and safety over the past two years. The best agent now completes nearly 90% of tasks and makes harmful mistakes just 2.5% of the time, down from 26%.
A satellite has autonomously located a target for the first time, marking a major milestone in AI-powered space technology. This breakthrough could revolutionize how we monitor Earth and respond to disasters.
The White House restricted exports of Anthropic's Mythos AI model partly due to concerns that a Chinese-linked group accessed it. This could have serious national security implications if true.
PyTorch has introduced a new optimization technique called fused MLP that speeds up AI model training. This advancement makes it easier and faster for developers to build and train complex AI models.
Notion has integrated OpenAI's Codex to help users quickly generate documents and code. This makes it easier for small teams to create and collaborate on projects without needing deep technical expertise.
Meta is reversing its $2 billion acquisition of AI startup Manus after Beijing ordered the deal undone. This highlights the growing tensions between global tech and Chinese regulations.