Skip to content
AI NEWS JOURNAL
AI, IN PERSPECTIVE

Search AI news

Search headlines, summaries and articles to connect the dots.

Newest first

The Silent Consensus

Dean Ball, Head of Strategic Futures at OpenAI, examines how more capable AI could reshape society and how people can retain control of their institutions.

Time-Series Models in the Streamhouse: A New Frontier and Use Cases

IBM's time-series foundation models, with over 50 million downloads, are now natively integrated into Confluent's platform, enabling real-time forecasting, anomaly detection, and optimization on streaming data using SQL. These models run on CPUs in both cloud and on-premises environments, supporting mission-critical applications across manufacturing, retail, banking, and airlines by delivering timely insights closer to decision-making points. The integration leverages Apache Flink and Confluent’s Streamhouse architecture to process data streams, apply AI-driven intelligence, and enable actions such as production optimization, fraud detection, supply chain planning, and trading execution—all in real time.

Token Commitment as a Control Surface for Steering Masked Diffusion Generation

arXiv:2611.21669v1 Announce Type: new Abstract: Masked diffusion models repeatedly predict provisional tokens before a sampler decides which predictions become permanent. Existing inference-time steering methods primarily modify activations, logits, or denoising trajectories, while treating this commitment policy as a fixed decoding detail. We instead study token commitment as a diffusion-specific control surface and separate it into two decisions: which tokens to commit and how many tokens to commit at each step. For the first, we introduce counterfactual target-aware commitment, which ranks unresolved positions by a mixture of model confidence and leave-one-token-out target contribution. For the second, we introduce calibrated adaptive budgets driven by target progress and confidence. Across language and vision-language settings in the LLaDA family, multiple targets, two activation-steering methods, and an unsteered control, both mechanisms yield empirically monotonic target-quality trade-offs in every configuration we evaluate. Adaptive schedules also recover competitive operating points relative to scorer-free stochastic schedules matched in average denoising cost. Because both mechanisms require only a scalar score evaluated on a provisional completion, they apply unchanged when the target is an automatic proxy for task success rather than an output attribute: NLI entailment for dialogue summarization and CLIPScore for captioning, neither of which corresponds to a single reusable steering direction. We reproduce the behavior of both mechanisms on a second masked-diffusion family, Dream-7B, for one control target and one task-success target. These results characterize commitment--which provisional predictions become irreversible, and how rapidly--as an interpretable control surface for steering masked diffusion generation under the conditions we test.

v1.20.0

Dify 1.20.0 introduces support for exporting traces to any OpenTelemetry collector, enables malware scanning of uploaded knowledge files, adds real-time Agent tool activity visibility in Chatflows, and allows automation via personal access tokens. New Human-in-the-Loop (HITL) features integrate approvals into messaging tools like Slack and Teams, while configuration changes include updated defaults for security and integration settings. The release also includes database migrations, improved performance, and bug fixes related to file handling, authentication, and worker stability.

Vercel Drives now reads files up to 6x faster

Vercel Drives has improved cold-read performance with a new read-ahead algorithm that anticipates data needs based on access patterns, reducing wait times without requiring SDK updates or configuration changes. Pricing remains unchanged.

What Does an Agent Need? Bundle-Aware Agent Resource Composition

arXiv:2611.22030v1 Announce Type: new Abstract: An agent resource's value depends on its backbone and companion resources and is known only after execution. Token-based generators remain tied to their training catalog, while independent resource scoring misses complementarities and reused bundles target other tasks. We formulate Task-Conditioned Agent Resource Composition (AgentComposition): jointly selecting a backbone model and a tool-skill bundle to maximize end-to-end execution utility. Our Retrieval-Grounded Bundle-Aware Agent Generation retrieves candidate resources and past configurations of similar tasks. A fine-tuned generator reads resource descriptions and recombines candidates through constrained beam search, allowing new resources without retraining. A bundle-aware critic ranks complete configurations using a large language model (LLM) teacher and historically successful configurations as execution proxies, and adapts to execution feedback when available. We evaluate reference recovery, offline execution with judged and native rewards, and online recommendation with real execution, building AgentSelect-Exec with 5,060 executed configurations. Our method recovers the complete reference configuration within ten candidates for 67.73% of AgentSelect queries, nearly twice the strongest baseline, and leads all baselines on SkillsBench. The proxy-trained critic raises the top-ranked agent's execution score from 7.58 to 8.37 out of 10; execution feedback increases it to 8.78, above the strongest baselines. Four target-domain queries suffice to adapt the generator to SkillsBench. Agents recommended online outperform the strongest baselines under both an LLM judge and native verifiers. These results support learning and transferring joint resource composition from task experience.

What To Measure When ARR Isn’t Enough

AI companies’ hybrid subscription, usage and outcome-based pricing makes headline ARR figures an unreliable measure of predictable revenue, often obscuring differences in customer commitments and spending patterns. Guest author Kenza Zayani, an early-stage investor at RTP Global, warns that as these businesses seek growth capital, investors increasingly expect evidence of retention, expansion, revenue predictability and sustainable margins to assess the durability of growth.

Your complete guide to AWS Quick at AWS re:Invent 2026

AWS Quick gives your whole organization AI that knows your business, respects your security rules, and acts on your behalf. At AWS re:Invent 2026, customers including Allianz, Cox Automotive, and Southwest Airlines share how they govern, extend, and scale Quick. Use this guide to find the sessions that fit your goals.

b12451

The web UI for Llama.cpp has been relocated from tools/ui to ui/apps/chat as part of a structural refactoring to separate the component library (ui/lib) from the applications built on it. The app is now officially named Llama Chat, with its package renamed to @ggml-org/llama-chat, while maintaining private status. Documentation, help text, comments, and other references have been updated to reflect the new name, though legacy identifiers remain unchanged for backward compatibility.

b12475

The Llama Chat UI has been upgraded to SvelteKit 3, integrating its built-in service worker for improved update detection and offline functionality. The migration includes updates to Vite 8, TypeScript 6, Storybook 10, and ESLint 10, along with the replacement of deprecated packages and resolution of npm audit issues through dependency overrides, including pinning minimatch to version 11.0.5 to address a security vulnerability. Changes also align with current Svelte patterns, such as using $state instead of stores and adopting new syntax for actions and class bindings.

Benchmarking LLM Inference on Desktop-Scale NVIDIA Grace Blackwell Clusters

arXiv:2611.20289v1 Announce Type: new Abstract: Two NVIDIA DGX Spark GB10 systems connect directly over 200 GbE to serve larger language models and increase inference capacity, but the benefits depend on how they are deployed. We benchmark 17 open-weight checkpoints spanning dense and mixture-of-experts architectures, 8B to 235B parameters and four weight precisions on one and two RDMA-connected systems. Using vLLM, we compare tensor parallelism (TP=2), pipeline parallelism (PP=2) and a two-replica baseline at concurrency 1 to 64, with three independent launches per configuration. TP=2 delivers a median 1.59x peak-throughput gain over one Spark, versus 1.39x for PP=2 and has higher throughput in 93 of 96 matched comparisons with separated 95% Welch intervals. It reduces unloaded inter-token latency to 0.52x to 0.73x of one Spark, whereas PP=2 provides little reduction. TP=2 also achieves the highest goodput under a strict 100 ms inter-token latency objective wherever both strategies are supported and the objective is met. For models fitting on one Spark, TP=2 generally favors low-concurrency serving, while replication offers higher throughput at high concurrency. One Spark uses less GPU energy per output token at matched concurrency. Across three precision ladders, NVFP4 reduces unloaded latency more than adding a second Spark at BF16. NCCL microbenchmarks and comparisons with an earlier TCP campaign suggest that collective communication overhead contributes to the advantage of TP=2. These results provide deployment guidance for desktop-scale inference.

Control where GitHub Copilot customizations come from

As an enterprise administrator, you can now require selected GitHub Copilot customizations to come from plugins or managed settings, rather than standalone user or project configuration. Combined with marketplace restrictions,… The post Control where GitHub Copilot customizations come from appeared first on The GitHub Blog.

Get answers from tables with tabular search in Amazon Bedrock Managed Knowledge Base

Amazon Bedrock Managed Knowledge Base now supports tabular search, so agentic retrieval can answer questions over CSV, XLSX, and XLS files. It looks up values and runs calculations across every relevant row, then combines the results with facts from related documents to answer multi-step analytical questions.

How ArisGlobal scaled generative AI inference on Amazon Bedrock

ArisGlobal runs NavaX, its life-sciences AI engine, on Amazon Bedrock. By caching the shared prompt prefix instead of each per-case document, ArisGlobal cut effective input cost by about 50% and overall monthly inference spend by about 40%, while onboarding more tenants without proportional cost growth.

How Duco transforms financial data automation with Amazon Bedrock AgentCore

Duco, a data automation platform for tier-one banks, rebuilt its agentic AI stack on Amazon Bedrock and Amazon Bedrock AgentCore. With strict per-tenant isolation, a multi-model strategy, and an MCP skills architecture that cut context overhead by 90%, Duco cut reconciliation deployment time 10x and onboarding costs 70%.

How the World’s Most Advanced AI Infrastructure Supply Chain Invents the Future

Deb Shoquist leads operations at NVIDIA, overseeing the supply chain and manufacturing behind its AI platforms. That description sums up the job, but it doesn’t begin to encompass what Shoquist has made possible over decades. Shoquist, her team and their global network of manufacturing partners enable NVIDIA to introduce world-leading technologies year-round. Much of the […]

Hy4 Flash now available on AI Gateway for 50% off

Tencent's Hy4 Flash model is now available on AI Gateway with a 50% discount on input, cached input, and output tokens until November 29, 2026. It offers the full feature set of the Hy4 family—reasoning with preserved thinking across turns, tool calling, structured outputs, prompt caching, image input, and a 1M-token context window—at approximately one-tenth the cost per token. The model can be used with the identifier 'tencent/hy4-flash' in applications and coding agents via Vercel's CLI setup. AI Gateway provides unified API access, usage tracking, cost management, and performance optimizations without markup or platform fees. Additional resources include reference architectures, agent stacks, web application guides, backend options, and a prompt library.