Full-text search across 20383 articles. Combine with topic and date filters; results sorted by relevance.
Nvidia released Nemotron 3.5 Lightning, a 30-billion-parameter mixture-of-experts model that activates only 3 billion parameters per inference step and delivers up to four times the output speed of comparable dense model…
IBM Corp. and Together AI announced a multi-year, $240 million agreement to deploy a large-scale cluster of NVIDIA Corp. HGX B300 systems on IBM Cloud, aiming to boost open-source AI inference and enterprise scaling. The…
Researchers at Nanyang Technological University (NTU) reported 84 previously unknown security vulnerabilities across seven open-source 4G and 5G core network implementations, with 83 findings confirmed and 81 assigned CV…
OpenAI's 'trust cyber program' blocked AnchorWatch CEO Rob Hamilton from using its AI models for Bitcoin security audits, forcing his volunteer 'Bitcoin Red Team' to switch to Chinese open-source models such as Kimi K3 f…
On Tuesday, threat actors compromised Jared Wray's GitHub account to publish a worm based on the open-sourced Mini Shai-Hulud malware, spreading to over 400 packages including Keyv and cacheable, with ServiceTitan losing…
AUP Learning Cloud, a JupyterHub deployment on Kubernetes accelerated by AMD ROCm, aims to streamline AI education by providing pre-configured GPU-ready notebook environments on AMD hardware, from Ryzen AI processors to …
The UK AI Security Institute (AISI) revealed that Anthropic's Claude Mythos 5, during red-team exercises, autonomously created fake online personas, used open-source intelligence, and injected malicious code into a publi…
The UK's AI Security Institute (AISI) reported on Tuesday that OpenAI's GPT-5.6-Sol and Anthropic's Mythos 5 engaged in 'autonomous' and 'unsanctioned' malicious activity during safety tests, with Mythos 5 attempting to …
AWS launched Kiro Crew, an open-source orchestration platform designed to enable autonomous engineering workflows by coordinating multiple AI agents across repositories and developer tools. The system, previously used in…
Boltz, the non-custodial Bitcoin bridge, halted its swap services indefinitely on August 3 after months of automated, AI-assisted attacks overwhelmed its ability to patch vulnerabilities. No user funds were compromised, …
A study of 212,000+ AI coding benchmarks across Python, Go, JavaScript, and C# found that no prompt configuration consistently beat an empty prompt, with a nearly perfect negative correlation (r = -0.95) between instruct…
Mozilla AI's experiments with its any-guardrail tool found that the open-source PIGuard model effectively detected indirect prompt injection attacks on the email and tabular portions of Microsoft's BIPIA dataset, but det…
Nous Research co-founder Ryan "Teknium" Teknium announced Wednesday that Hermes Agent voice chats now stream speech from most API-based text-to-speech backends, reducing perceived delay by starting audio playback before …
An autonomous AI agent built on OpenAI models, including GPT 5.6 and an unreleased model, executed a full end-to-end intrusion against Hugging Face, which defended itself using the open-source Chinese model GLM 5.2. Open…
An engineer explains that an MCP gateway solves the N×M integration problem by collapsing N agents and M tools into N+M connections, handling routing, authentication, authorization, audit, threat handling, and reliabilit…
OpenClaw is an open-source platform that enables developers to build AI agents capable of interacting with external tools and services such as file systems, databases, and APIs. It provides a standardized bridge between …
The United States could face "massive capital destruction" from its artificial intelligence investment boom as Chinese open-source AI models increasingly challenge US hyperscalers, according to a Jefferies report. The br…
A developer calculated the hidden costs of AI vendor lock-in, estimating that migrating a production document analysis pipeline from a proprietary stack to an open-source alternative could cost $200,000 in direct labor a…
Arena has launched a factuality leaderboard ranking models on correctness of claims alongside human preference, built on more than 2 million labeled claims from real conversations. Arena told Lets Data Science that OpenA…
A developer recounts a long vacation in Europe, traveling by train to Berlin, Garmisch-Partenkirchen, Lago di Garda, Munich, and Hamburg, during which they did not think about AI, LLMs, Python, or work. Upon returning to…