ZeroHour
DashboardNewsCVEsExploitsIncidentsAIBriefAnalyticsEntitiesWatchlistIOCsAsk the feedSourcesFeeds & APIKiosk mode

Search: “parallels”

8 stories in the last 24h

Searchkeywordsmart Sortlatesthotimportancerelevance
DomainAllCyberAIWindow24h3d7d30dAllIn the wildImportant onlyShow duplicates
CategoryAllAI industry 3AI research 3AI tools & infra 2
SeverityAllcritical 0high 0medium 0low 0info 8

Anthropic Launches Claude Code Projects in Beta: Parallel Cloud Sessions That Keep Running After You Close Your Laptop

Anthropic launched Claude Code Projects in beta, letting one conversation spawn parallel cloud sessions on separate branches that continue after logout.

Anthropic redesigned Projects in Claude Code so a single coordinator conversation spawns parallel threads, each a full Claude Code cloud session with its own git branch and repository copy. Threads inherit project instructions (up to 16,000 characters), MEMORY.md project memory, skills, plugins, and repositories, and can further delegate via subagents, loops, and workflows. The beta is limited to select Pro and Max users on web and desktop, with a cap of 200 new threads per day and faster consumption of plan limits since every running thread is a full session.

MarkTechPost · 17h agoAI tools & infra

Bend – A language that blocks AI mistakes via proof and runs on GPUs

Bend is a new programming language combining C-speed compilation, GPU parallelism, and Lean-style proofs that stop AI agents from merging code violating declared laws.

Bend is a programming language whose type checker acts as a proof checker, letting developers declare invariants in LAWS.bend that AI coding agents must prove with PROOF.bend before committing. It compiles to native code running near C speed on a single core and up to 100x faster across GPU cores, with automatic parallelism requiring no threads or locks. The project publishes two papers, BendTT (an affine dependent type theory) and BendRT (a parallel CPU/GPU runtime), and integrates with AGENTS.md workflows for AI-driven 'vibe coding'.

Hacker News · AI · 17h agoAI tools & infraHN 153↑ · 68 comments

Anthropic keeps pushing Claude Code toward autonomous coding with new parallel agent workflows

Anthropic rebuilt Claude Code's Projects feature to split goals across parallel cloud agent threads that can open pull requests and run tests.

Anthropic's updated Claude Code Projects feature uses a coordinator that splits a user's goal into parallel threads, each running as its own cloud session, with shared memory and a library of uploaded files and results. Threads can open pull requests and run tests, and progress is trackable in the main chat or per thread, including on mobile. The beta is open to select Pro and Max subscribers, with Team and Enterprise access and local execution to follow. The update follows Anthropic making autopilot mode the default in Claude Code.

The Decoder · 19h agoAI industry

Claude Code relaunches Projects to manage multiple AI agents in the cloud

Anthropic relaunched Claude Code Projects, letting users run multiple parallel Claude Code cloud agents with shared memory, goals, and files.

The revamped Claude Code Projects feature organizes work as "threads," each a Claude Code cloud session working on its own branch and copy of the repo, with a coordinator directing tasks; overlaps on the same code resolve as merge conflicts. Threads can further split work using subagents, loops, and workflows, and users can interact per-thread or via the main project chat. The feature launches today in beta for select Claude Pro and Max subscribers, with later expansion to all Pro, Max, Team, and Enterprise users plus Cowork and regular Claude chats. Local tool and code support is promised "very soon."

The Verge · AI · 19h agoAI industry1

[AINews] not much happened today

Latent Space AI news digest covers Anthropic's Claude Code Projects, Google's managed agent APIs, TypeSafe's Jev classifier, and OpenAI's Astra for Law launch.

The 9/16-9/17/2026 AI news roundup highlights Anthropic's Claude Code Projects enabling one conversation to spawn parallel cloud sessions, and Google's Gemini managed agents adding a Credentials API, Files API, and claims of 30% lower costs. It also covers TypeSafe's Jev, a fast constrained-output classifier being used for routing, judgment, and structured decisions, with open reproductions such as openjev-s on Qwen3.6-35B-A3B. OpenAI launched Astra for Law with 26 partner-built and 47 community plugins via Trusted Access, with reports it beats generic GPT-6 Astra plus web search on Vals' legal benchmark. Research items include DeepMind's Stellar Colosseum multi-agent math harness (Codeforces 4263, 71.0% on TCS-Bench) and NVIDIA-associated Agora using Git commits as shared memory.

Latent Space · 7h agoAI industry 2 sources

Unifying Models of Intergroup Hostility in Online Discourse

Researchers unify six theories of intergroup hostility using 2.86 million TikTok, Truth Social, and Twitter/X posts, finding boundary and threat construction anchor rhetoric.

The study models mechanisms from six foundational theories of intergroup hostility - boundary construction, threat construction, scapegoating, negative evaluation, dehumanization, and action orientation - within a common empirical framework using 2.86 million posts from TikTok, Truth Social, and Twitter/X during the 2024 U.S. presidential election. Structurally, boundary and threat construction anchor the system; temporally, boundary construction, derogation, and action orientation appear early, dehumanization and threat construction later, and scapegoating last. The work aims to give computational social science a unified empirical basis for modeling hostile rhetoric beyond single-label detection.

arXiv cs.AI / cs.LG / cs.CL · 20h agoAI research

Large Language Models as Falsifiers for Cyber-Physical Systems

LLM-Falsifier uses large language models with semantic prompting to minimize STL robustness and find counterexamples in cyber-physical systems more efficiently.

The paper formulates falsification of Signal Temporal Logic specifications as robustness minimization and leverages iteratively prompted LLMs as optimizers. It exposes the LLM to semantic information absent from numerical optimizers, including natural-language signal names, output trajectories, and critical-time witnesses for the minimum robustness value. On ARCH-COMP falsification benchmarks, LLM-Falsifier required fewer simulations than surrogate-based, Bayesian, and search-based tools on 14 of 21 specifications.

arXiv cs.AI / cs.LG / cs.CL · 20h agoAI research

dQwen3.5: Hybrid-Attention Diffusion Language Models

Researchers adapt Qwen3.5 hybrid-attention backbones (0.8B-9B) into diffusion language models, reaching training loss in half the tokens versus full attention.

Researchers converted pretrained Qwen3.5 models at 0.8B, 2B, 4B, and 9B scales into the dQwen3.5 family of diffusion language models. The work addresses hybrid attention-RNN backbones, whose RNN layers are structurally causal and hard to bidirectionalize. Against a full-attention control, the hybrid backbone reaches a given training loss in roughly half the tokens. The resulting models match full-attention DLMs in any-order decoding and perform strongly under parallel decoding.

arXiv cs.AI / cs.LG / cs.CL · 20h agoAI research