The Law Society Tribunal (Ontario) ordered suspended lawyer Shahryar Mazaheri to pay $31,150 in costs after he filed motion materials drafted with generative AI that cited fabricated and irrelevant case law. The decision emphasizes the need for human review and oversight of AI outputs.
Australia’s eSafety Commissioner contacted OurDream and is considering its compliance with Online Safety Act codes and standards, including obligations related to child sexual exploitation material and age-restricted pornography. The regulator may pursue further action, including civil penalties.
South Korea’s Financial Services Commission and the Ministry of Trade, Industry and Resources announced policy financing of about 16 trillion won (~$10.3B) to accelerate adoption of physical AI via the National Growth Fund–M.AX Frontier Project.
Illinois laws take effect July 1, 2026, including House Bill 3851 making posting unauthorized AI images cyberbullying in schools. The article also notes other new Illinois laws, including prediction market regulation and related litigation.
The U.S. Department of Commerce reversed an export ban on Anthropic’s Claude Fable 5 and Claude Mythos 5, lifting export controls for Fable while access to Mythos remains limited for some partners. The article describes the change as a withdrawal of the prior license requirement.
Autonomous AI Research promises to accelerate the scientific progress of machine learning. To realise this goal, current Large Language Model (LLM)-based agents need to go beyond just writing code, to mastering the exploration of simultaneously performant, diverse and novel ideas. To this end, we introduce Heuresis, a framework that abstracts the research pipeline into a set of general and composa
AI agents that pay for resources via the x402 protocol embed payment metadata - resource URLs, descriptions, and reason strings - in every HTTP payment request. This metadata is transmitted to the payment server and to the centralised facilitator API before any on-chain settlement occurs; neither party is typically bound by a data processing agreement. We present presidio-hardened-x402, the first
Detecting fraudulent credit card transactions remains a significant challenge, due to the extreme class imbalance in real-world data and the often subtle patterns that separate fraud from legitimate activity. Existing research commonly attempts to address this by generating synthetic samples for the minority class using approaches such as GANs, VAEs (Variational Autoencoders), or hybrid generative
End-to-End autonomous driving (E2E-AD) systems face challenges in lifelong learning, including catastrophic forgetting, difficulty in knowledge transfer across diverse scenarios, and spurious correlations between unobservable confounders and true driving intents. To address these issues, we propose DeLL, a Deconfounded Lifelong Learning framework that integrates a Dirichlet process mixture model (
Quantization-aware training (QAT) is essential for extremely low-bit large language models (LLMs). Current QAT methods are mainly based on scalar quantization (SQ), which enables efficient optimization but suffers from severe performance degradation at 2-bit precision. On the other hand, vector quantization (VQ) provides substantially higher representational capacity, but its discrete codebook loo
Capital is concentrating in large, late-stage-style checks rather than a broad spread of small rounds. The pattern points to money flowing toward AI infrastructure and hardware-linked platforms, with one named AI company drawing strategic backers.
Capital is concentrating in large, late-stage-style checks rather than a broad spread of small rounds. The pattern points to money flowing toward AI infrastructure and hardware-linked platforms, with one named AI company drawing strategic backers.
Capital is concentrating in large, late-stage-style checks rather than a broad spread of small rounds. The pattern points to money flowing toward AI infrastructure and hardware-linked platforms, with one named AI company drawing strategic backers.
Named hires continue to cluster around senior AI leadership and research authority, signaling where teams are trying to deepen technical credibility. The moves also show established companies and new labs competing for the same high-status operators.
Named hires continue to cluster around senior AI leadership and research authority, signaling where teams are trying to deepen technical credibility. The moves also show established companies and new labs competing for the same high-status operators.
Named hires continue to cluster around senior AI leadership and research authority, signaling where teams are trying to deepen technical credibility. The moves also show established companies and new labs competing for the same high-status operators.
The benchmark desk is thin but still useful as a snapshot of where baseline capability is being logged. The entries cluster around general reasoning and math, suggesting attention remains on broad competence rather than narrow task spikes.
The benchmark desk is thin but still useful as a snapshot of where baseline capability is being logged. The entries cluster around general reasoning and math, suggesting attention remains on broad competence rather than narrow task spikes.
The benchmark desk is thin but still useful as a snapshot of where baseline capability is being logged. The entries cluster around general reasoning and math, suggesting attention remains on broad competence rather than narrow task spikes.