Reading a tensor value into Python can split compiled model code and force the processor to wait. IBM explains the two separate costs.

Browserbase separates reliable cloud browsers from adaptive page control, with receipt collection, vendor payments and business-verification examples.

SABench tests whether an assistant finishes the requested work and whether every write stays within permission. Replying correctly to an email can coexist with an unrequested change to a linked ticket. The study uses 201 synthetic office …

Kepler separates language-model planning from financial facts and arithmetic, then checks every displayed number against a citation registry.

Kaggle's benchmark audits expose fragile scores. LinqAlpha's MandateBench tests the reasoning behind a full investment memo, with early results from ten tasks.

Memory Tax compares eleven memory systems with search over the original conversation. It asks whether extracting and organizing memories earns its extra cost. The authors separate new answers that memory enables from answers lost relative …

GEPA evolves prompts for model programs, including workflows built with DSPy. This study tests its merge operation, which combines candidates descended from a common starting instruction. The comparison spans five datasets, two models and …

Creative-industry employment is shrinking. Politano examines how generative AI could weaken the businesses that pay artists, while leaving its share of the losses unresolved.

Laurie Voss argues that faster coding moves the bottleneck toward defining what users need. His product-engineer forecast also demands a new apprenticeship for junior developers.

The Brotli patch extends beyond Node.js Python web clients also need fixes for the Brotli flaw — Updating Node.js alone leaves other copies of the vulnerable decompression library in download tools, application containers and proxies. The …

Sofia Jakovcevic builds a regression-testing loop from production traces. Keep separate scores for safety and task completion, and protect workflows that already work.

Bridgewater's Vault combines governed documents, measured search quality and familiar research workflows. Sam Green and Aaron Linsky explain what people and agents need from it.

Personal data shared for AI answers can also inform ads Google lets connected Gmail and Photos data influence its ads — Connecting personal content for better AI answers can also permit advertising uses, making the receiving service’s …

Harneet Kaur tests two models on iHeartMedia. Checking each reasoning layer makes their forecasts converge, but agreement leaves accuracy unproven.

Contour's Hari Kumar compares research workflows with and without AI, tracks cost per useful insight and counts the time lost fixing broken tools.

Jim Nielsen compares AI-lab conviction with Cold War nuclear planners. The HN discussion asks what evidence supports the analogy and who would actually share the promised abundance.

SE Gyges traces the Berkeley network behind existential-risk advocacy, arguing that shared funding, careers and intimate relationships deserve public scrutiny. The evidence supports narrower claims than the headline.

Kimi releases a new hosted model, with downloads promised later Moonshot releases Kimi K3.1 Pro and a faster variant — Independent measurements make the new model worth comparing on coding and document work, while local deployment must wait …

Aman Singh explains how Arca coordinates days-long client work with event signals, persistent memory, shared plans and human approvals.

Andrey Rybka maps agent incidents to scoped identities, checked tool calls, tenant isolation and outbound controls that include DNS.

Matt Wolfe’s weekly roundup spans Claude Duet, software-security incidents, editable vectors, agent access and cautiously framed medical-AI results.

Jump Trading’s compute leads explain bespoke GPU contracts, startup prepayments and why a price hedge cannot deliver a cluster.

Separate model sessions recover messages from rendered drawings, while the decoding tools leave a visible trail.

Joseph Kim describes licensed financial data, firm templates and inspectable citations, then separates today’s product from his 2027 predictions.

Hard Fork's Rob Copeland and Erin Griffith connect debt-funded computing capacity to the limits of what AI customers will pay.

Carmen Li explains reserved GPU procurement, provider verification and forward pricing, separating delivered capacity from financial hedges.

Cole Medin adds shared ownership, context, readiness checks and human approvals around his coding-agent loop in a sponsored Port walkthrough.

MCP-Injector inserts errors, delays and altered results between an assistant and tools connected through MCP, the standard for connecting agents to tools. It tests reliability when connected programs behave badly. The authors cross …

Structured output can still be wrong Minimum list lengths push models to pad extracted answers paper notes → — A valid output format can conceal incorrect data when the document contains fewer answers than your schema demands. The October …

IBM’s panel weighs distributed-model security, clinicians’ use of OpenEvidence and the limits of letting agents coordinate with one another.

Freddy Lecue describes Wells Fargo’s transaction-trained transformer, its data-tokenization work and reported fraud gains over the bank’s XGBoost baseline.

Isabelle Williams demonstrates agent registration with auth.md, user-approved identity sharing, and separate revocable agent tokens with limited permissions.

Stephanie Jarmak explains how realistic workloads, repeated runs, trusted traces and enforced quality gates make agent tests useful for real decisions.

Anthropic and Theorem report checked proofs for 239 test entry points across 123 libraries. The public release, machine model and patching process need separate scrutiny.

ModelCompare's small experiment finds a familiar refrain about tribal loyalty and rationalization. Its broad answer categories and leading prompt limit what that agreement means.

San Francisco Compute’s Evan Conrad argues that deposits, delivery delays and uncertain demand form one financing problem. Tradable capacity needs reliable cluster operations behind it.

Outside reviewers find gaps in OpenAI’s monitoring Outside reviewers found ways for OpenAI’s assistants to avoid monitoring — The newly published review shows why recording actions and independently checking that recording happened belong …

A refund request can commit even when its acknowledgement disappears. If permission changes before a workflow restarts, a retry needs two separate checks: current authority and evidence about the earlier effect. Arya.ai studies 336 …

ImpasseBench tests whether coding assistants stop when their own tool results show that outside help is needed. It separates persistent barriers from temporary errors and blocked testing. The authors run eight model configurations on 102 …

Upwork’s SWE-Clar asks coding assistants to clarify incomplete requests before building software. It freezes that conversation, then starts two isolated continuations: one answers requirement questions and the other writes code. Assessment …

Jellyfish’s Nicholas Arcolano finds that skills, usage, team environment and individual differences explain more variation in merged code changes than tool choice. Quality remains outside this analysis.

Peter Yang turns account cleanup into a read-only AI review: inspect access, passwords, sign-ins, API keys, updates and suspicious messages, then approve specific fixes.

A coding assistant can preserve its task score while forgetting restrictions on how it should work. This study compares five conversation-shortening strategies inside OpenCode, the software running the model’s tools and commands. The …

MacPaw researchers test four families of model-serving optimizations on Apple Silicon. Their measurements show why a server-GPU speedup needs checking against the model, software and request pattern you actually use. Pruning the KV cache, a …

AfterDeleteBench separates deleting a record, preventing disclosure and blocking a later write. Its controlled tests also expose failures hidden by answer parsers.

DoorDash describes a narrow coding task with explicit safety checks: remove obsolete feature flags, the settings used for gradual releases and experiments. The system queries deployed values before an engineer approves each proposed …

A small model learns terminal behavior, but removing one layer preserves plausible output while destroying its memory.

Omnara moves its agents off the laptop Omnara introduces a self-hostable service for long-running assistants — Its new platform separates the running assistant from the machines it controls, so closing a laptop need not stop the work. An …

Grace Ableidinger follows a model request through tokenization, routing, batching, prefill and decode, and explains what first-token and inter-token latency measure.

Spencer McIntyre on AI-assisted security research, broken proofs of concept, and why local models appeal to teams handling unpublished vulnerabilities.