AI News - Live AI News & Trends | Blockchain.News
AI News
|
OpenAI GPT5.6 Public Launch, ChatGPT Work Debuts
According to TheRundownAI, OpenAI launched GPT-5.6 publicly and introduced ChatGPT Work, while Meta shipped Muse Spark 1.1 and new tools emerged. (Source) More from The Rundown AI 07-10-2026 02:30 UTC |
|
GPT5.6 Boosts Copilot Apps Performance
According to gdb, GPT-5.6 with Work IQ rolls out to Copilot Chat, Cowork, M365, GitHub, and Foundry, improving reasoning and content quality. (Source) More from Greg Brockman 07-09-2026 22:49 UTC |
|
GPT5.6 Sol Builds No Code City Demo
According to EthanMollick, GPT5.6 Sol in Codex recreated a procedural brutalist city builder without code in under a year of progress. (Source) More from Ethan Mollick 07-09-2026 19:46 UTC |
|
Tesla Robotaxis Face New Jersey Sensor Bill
According to Sawyer Merritt, a NJ bill mandating cameras plus LiDAR and radar could block Tesla robotaxis; Uber urged keeping human drivers. (Source) More from Sawyer Merritt 07-09-2026 19:23 UTC |
|
Tesla FSD v14 Lite Expands Internationally
According to SawyerMerritt, Tesla begins rolling out FSD v14 Lite to US-built Model 3/Y with HW3 in South Korea, marking its first international release. (Source) More from Sawyer Merritt 07-09-2026 17:42 UTC |
|
TRACE Boosts Qwen3.6-27B to 73.2% on SWE-bench
According to StanfordAILab, TRACE trains agents on missing skills, pushing Qwen3.6-27B to 73.2% on SWE-bench Verified with under a quarter rollouts. (Source) More from Stanford AI Lab 07-09-2026 15:45 UTC |
|
LLM-as-a-Verifier Delivers SOTA Across 4 Benchmarks
According to StanfordAILab, LLM-as-a-Verifier scales verification to SOTA on Terminal-Bench V2, SWE-Bench Verified, RoboRewardBench, MedAgentBench. (Source) More from Stanford AI Lab 07-09-2026 15:45 UTC |
|
Distill to Detect exposes hidden LLM bias
According to StanfordAI Lab, D2D amplifies subtle fine tuning shifts into text to reveal hidden LLM bias for auditors. (Source) More from Stanford AI Lab 07-09-2026 15:30 UTC |
|
Language Models Fail from Data Repetition
According to StanfordAILab, an ICML 2026 workshop oral shows internal data repetition degrades language models, earning runner-up recognition. (Source) More from Stanford AI Lab 07-09-2026 15:29 UTC |
|
Federal Reserve Task Force embraces AI for risk oversight
According to @CNBC, new Fed task force members align with Chairman Kevin Warsh’s AI push to enhance supervision, stress testing, and productivity analytics. (Source) More from CNBC 07-09-2026 15:03 UTC |
|
Meta Llama upgrades challenge OpenAI, Anthropic
According to @CNBC, Meta ships a major Llama update to rival OpenAI and Anthropic, targeting enterprise use, safety, and multimodal features. (Source) More from CNBC 07-09-2026 14:29 UTC |
|
ChatGPT Work vs Codex Analysis clarifies roles
According to @emollick, users are unclear on ChatGPT Work versus Codex tradeoffs, prompting security and capability comparisons, per X post and vendor docs. (Source) More from Ethan Mollick 07-09-2026 14:25 UTC |
|
SpaceX Starmind AI1 unveils 150 kW breakthrough
According to SawyerMerritt, SpaceX revealed AI1 specs with 150 kW compute, vendor-agnostic chips, laser links via Starlink, and Bastrop-built solar tech. (Source) More from Sawyer Merritt 07-09-2026 13:11 UTC |
|
SpaceX Starmind unveils AI1 satellite details
According to SawyerMerritt, SpaceX launched a Starmind site detailing AI1 power, thermals, and solar, after an FCC filing to deploy 1 million AI satellites. (Source) More from Sawyer Merritt 07-09-2026 13:07 UTC |
|
GPT5.6 Boosts Work IQ across Copilot
According to satyanadella, GPT-5.6 with Work IQ lands in Copilot Chat, M365, GitHub, and Foundry, improving reasoning and output quality without speed loss. (Source) More from Satya Nadella 07-09-2026 12:35 UTC |
|
GPT56 Powers Broccoli Farm Workflow
According to OpenAI... A broccoli farmer named Hiroki runs operations with GPT-5.6, showcasing AI-driven scheduling, monitoring, and task automation. (Source) More from OpenAI 07-09-2026 12:04 UTC |
|
Meta Muse Spark 1.1 Earns Early Praise
According to AIatMeta, leaders praised Muse Spark 1.1 for faster multimodal generation and workflow boosts, citing early-access feedback on July 9, 2026. (Source) More from AI at Meta 07-09-2026 11:53 UTC |
|
Codex 5.6 Sol Accelerates Book Edits in 30 Minutes
According to emollick, Codex with 5.6 Sol reviewed a full book PDF in 30 minutes and delivered dozens of accurate editorial notes. (Source) More from Ethan Mollick 07-09-2026 11:00 UTC |
|
OpenAI Codex Powers ChatGPT Work Rollout
According to sama, OpenAI unveiled ChatGPT Work using Codex and GPT5.6 to act across apps, sustain long tasks, and deliver finished work. (Source) More from Sam Altman 07-09-2026 10:59 UTC |
|
GPT5.6 Triggers AI price war as costs plunge
According to God of Prompt, GPT5.6 sparked a price war: Luna $1 per 1M tokens, Sonnet 5 intro $2, Terra 2x cheaper than GPT5.5. (Source) More from God of Prompt 07-09-2026 10:58 UTC |