Newsletter ·
Daily 'AI for Work' Pulse: 17th of October
Ease into your Sunday with 14 new AI tools and 53 AI news articles. Plenty of fresh finds to skim while you unwind-pick a couple standouts and head into the week ready.
Jeroen Erne
Latest AI Tools
Navige
Navige routes AI agent requests to the smallest suitable model to reduce costs without changing the original model. It is built for engineering teams who want per-task and per-agent cost visibility, budgets, and audit trails.
Read more →
withHuman
withHuman acts as a gatekeeper between your AI agents and their tools, intercepting tool calls before they execute. It automatically approves routine actions based on your rules and escalates risky ones for human approval. Built for teams that alr...
Read more →
Actually Asked
Actually Asked embeds a FAQ section into your site using your own design. Visitors ask questions that are answered from your pages, and unanswered queries are forwarded to you. It is for site owners who want a self-improving FAQ that captures real...
Read more →
SuperLocalMemory
SuperLocalMemory is a local-first memory system for AI coding agents like Claude Code, Codex, Cursor, and ChatGPT. It lets you search images and PDFs by describing them and always indicates whether a memory is supported, related, or unavailable. B...
Read more →
Afterparty
Afterparty turns meeting transcripts into ready-to-review drafts of follow-up emails, Slack messages, CRM notes, and calendar invites. It is for anyone who wants to save time by having action items drafted and routed to the right tool for approval.
Read more →
Shrine
Shrine is a Mac menu bar utility that tracks the status of multiple coding agents so you can see which one needs your attention. It is built for developers managing concurrent agent sessions and lets you collect in-app assets as agents complete ta...
Read more →
Dreamscale Labs
Dreamscale Labs provides a cloud inference API for running robot AI models with real-time latency. It lets deployment and research teams test the latest physical AI models without needing to quantize them for small on-board GPUs.
Read more →
Superdraft
Superdraft learns your brand to generate designed email drafts with copy, images, and layout. It helps founders and marketers turn product updates or events into campaigns they can edit and export.
Read more →
Atticus
Atticus is a Mac assistant activated by holding Control-Option-Space that executes spoken commands like opening apps, filling forms, or completing purchases. It reads your screen to answer questions about open documents and emails, asking permissi...
Read more →
CmdRound
CmdRound is a hiring assessment that lets candidates complete a coding task with an AI agent. Recruiters receive a replay of every prompt and code change along with a score on verification, recovery, and direction skills.
Read more →
Rushilo
Rushilo is a desktop video editor that lets you control cuts, captions, and timeline edits using your own AI coding agent like Claude Code or Codex. It is built for creators who want agent-driven editing with full manual control, local file proces...
Read more →
Brussle
Brussle calibrates a correction to your decision model's output and applies it only when it improves outcomes on unseen data. It re-runs decisions when source data changes and refits the correction when the model's behavior changes. The tool is bu...
Read more →
Audimate
Audimate is an AI assistant that works directly inside your existing apps. Point at elements on your screen and describe what you want it to do, so it can guide you through tasks or run them automatically. It is built for professionals who want to...
Read more →
Lleverage
Lleverage provides AI agents that automatically process orders, invoices, and purchase orders inside existing ERP systems. It is built for operations teams in manufacturing and distribution who want to reduce manual data entry and only handle exce...
Read more →
All AI News from Yesterday
53 News Articles published.
Latest General AI News
Latest AI News for Creatives
Epic deploys vibe coding to rebuild its game launcher from scratch
Epic Games shipped a rebuilt Games Launcher coded mostly by AI, claiming it loads faster than Steam. The company says the AI-first rewrite let a small team ship a cross-platform client from scratch, with cold starts up to 5x quicker.
Read more →
Simon Willison builds tree appreciation game with Claude after three-hour AI coding session
Simon Willison built a browser game called Tree Appreciator using a single prompt to Claude Opus 5.5 Max, which took 3 hours and 7 minutes to generate the full experience with music, animation, and achievements.
Read more →
NetShort plans external access to AI short-drama tools
NetShort will let outside creators use its internal AI tools to make short dramas, while keeping a focus on rights, quality, and transparency.
Read more →
YouTube adds AI labels to music in YouTube Music and main app
YouTube is rolling out AI labels on songs across YouTube Music, flagging tracks detected as primarily generative AI-made. The move follows an AI remix topping Australian charts and industry mandates that chart-eligible music be substantially human-made.
Read more →
Google shares 10 research breakthroughs that amplify human ingenuity
Google demonstrated 10 research projects moving AI into real-world workflows, including a geospatial reasoning API and a $10 million crisis-resilience fund. MedGemma, an open health AI model, has been downloaded more than 10 million times by over 65,000 developers.
Read more →
Figma launches Specs editor for AI design rules
Figma launched Specs, a Markdown editor that stores design rules, code references, and guidelines in one file for both people and automated tools.
Read more →
Amazon's Bee launches Frontier and Eclipse AI recorders
Amazon’s new AI wearables record and summarize conversations, raising workplace concerns about consent, data security, and recording policies.
Read more →
OpenAI rolls out GPT-Live-1.1 with 13 new voices
OpenAI launched a voice update with 13 new voices and better accents, languages, and control.
Read more →
3DCodeVerse releases 2.03 million executable 3D code pairs across six languages, achieves 90% human-preference win rate over Claude design
A new open-source dataset delivers 2.03 million executable 3D code pairs and a verification harness that achieves a 90% human-preference win rate against Claude Opus 5.5 Design.
Read more →
Creative groups launch charter for ethical generative AI use
Creator groups demand enforceable rules for how AI uses creative work, including consent and disclosure.
Read more →
Latest AI News for Customer Support
TELUS Digital will implement Parloa customer-service agents
TELUS Digital and Parloa are teaming up to sell automated customer conversation software with managed services, so buyers must check how the two vendors split security, handoffs, AI models, and ongoing operations.
Read more →
WeInfuse and Clarion connect AI agents to infusion workflows
WeInfuse and Clarion linked AI agents to infusion center systems so they handle routine patient calls using live data and route complex cases to staff.
Read more →
DriveCentric introduces customer data platform for dealer groups
DriveCentric launched a customer data platform for dealer groups that unifies identity, sales, service, and communications records while keeping existing permissions intact.
Read more →
Specialty1 Partners invests in AI for dental practice operations
Specialty1 Partners is using AI in its dental practices for tasks like clinical notes, patient messages, and insurance work, while stressing the need for privacy, accuracy, and human oversight.
Read more →
Italian bank loses $108 million in AI voice and messaging fraud targeting senior executives
A fraudster used AI-cloned voices to trick an Italian bank chairman into transferring €95 million overseas. Authorities recovered roughly €53 million, while the rest was converted into cryptocurrency.
Read more →
Mews survey finds travelers prefer AI for hotel rate tracking
A survey of 3,250 travelers found they prefer AI tools for tracking hotel prices and translating languages over automated itinerary planning.
Read more →
EU survey finds 42% of smaller companies use AI or automated management
A European Union survey found that 42% of smaller companies use AI or automated management tools, mainly for customer service, performance monitoring, and scheduling.
Read more →
Nabla integrates OpenEvidence into clinical AI assistant
Nabla’s AI assistant now lets clinicians ask patient-specific questions and get answers linked to medical literature, requiring anonymization and citation checks.
Read more →
Moomoo adds AI assistants to global paper-trading contest
Moomoo added four AI research assistants to its paper-trading contest to help users test strategies with virtual capital.
Read more →
Latest AI News for Education
Student-built campus-events app wins Stellic AI challenge
A KU student won an international contest by creating a campus event app using AI tools provided in the challenge.
Read more →
AIPOCH releases an open-science AI research workbench
AIPOCH released an open-source AI research workbench that combines AI agents, literature, data, and code execution to support reproducible research.
Read more →
Hugging Face releases Llama Windows for local AI models
Hugging Face launched Llama Windows, an app that lets users run AI models locally without admin rights, boosting privacy but lacking built-in governance controls.
Read more →
Google gives 200 European scientists access to AI models
Google gives 200 European scientists access to its AI models and $3 million in computing credits, with transparent selection and independent publication required.
Read more →
CERN and Mistral plan foundation model for collider physics
CERN and Mistral are building an AI model trained on simulated particle collision data to improve event reconstruction and anomaly detection.
Read more →
Elsevier pledges 4,000 AI tool licenses for European researchers
Elsevier is giving 4,000 AI tool licenses to European researchers and expects universities to track usage and research impact.
Read more →
Study tracks 270 million clinical AI queries on OpenEvidence
A study of 270 million AI queries from 950,000 clinicians shows heavy use for treatment decisions, highlighting the need for quality evidence and more outcome research.
Read more →
OpenAI offers US university labs free GPT-Rosalind access
OpenAI is giving university biology labs free one-year access to GPT-Rosalind for up to 30 researchers, while requiring them to manage data rules, reproducibility, and publication policies.
Read more →
Neural networks are already clean and don't need untangling, researcher argues
A new analysis claims superposition-the idea that AI neurons pack unrelated concepts together-was misread, and that most neuron mixtures resolve into single concepts like a "V-shape" detector instead of "cat and car." The researcher estimates 90% of neuron poles in a 32B-parameter model have one...
Read more →
Latest AI News for Executives and Strategy
EQT AI Infrastructure fund reaches $11.5B in net asset value
EQT’s AI infrastructure fund grew to an $11.5 billion net asset value, driven by inflows and portfolio gains tied to AI expansion.
Read more →
Anthropic gives METR employee-level access for safety reviews
Anthropic lets six external evaluators access some AI systems to publish safety findings without company veto, while keeping customer data private.
Read more →
Anthropic reports cash-flow-positive quarter
Anthropic had a cash-flow-positive quarter, signaling AI can make money, but growth is slowing.
Read more →
Nvidia CEO says chip supply could stop over unsafe AI
Nvidia’s CEO said the company might cut off chip supplies to customers building unsafe AI, using hardware access as a way to enforce safety.
Read more →
Microsoft completes pre-training of MAI-Base-2, a 2.8 trillion parameter foundation model
Microsoft trained MAI-Base-2, an 80B-active-parameter model, on 16,000 GPUs over 38 days and claims 4-12x better pre-training efficiency than DeepSeek-V4-Pro-Base.
Read more →
Bank of America hires 1,000 contractors to patch software flaws exposed by Anthropic's Mythos
Bank of America hired 1,000 contractors after Anthropic's Mythos AI model exposed software flaws, driving remediation costs up by a couple hundred million dollars and sending its stock down 8%.
Read more →
Singtel Digital InfraCo recognized for integrated AI infrastructure
Singtel Digital InfraCo was recognized for building integrated AI infrastructure across Southeast Asia and advises buyers to evaluate the full stack—not just GPUs—by considering workload placement, data sovereignty, resilience, and total cost.
Read more →
Latest AI News for Finance
American Express introduces intent checks for AI-agent purchases
AmEx proposes a system where AI agents make payments using pre-approved limits and tokens, without ever seeing your actual card details.
Read more →
Latest AI News for Government
GUDEA uses AI to track online narratives before crises spread
GUDEA uses AI to map how stories and rumors spread across websites, helping teams spot emerging issues early, but it does not verify the information.
Read more →
UK medicines regulator plans AI safety tools by 2030
The UK’s medicines regulator plans to use AI tools for safety checks and trial reviews by 2030, and will set clearer rules for AI medical products.
Read more →
Latest AI News for Healthcare
Illumina launches BioInsight platform for genomic research
Illumina launched a platform that combines genomic tools, AI models, and biological datasets in one place with usage-based access.
Read more →
PhaseV validates GLP-1 adverse-event model on patient data
PhaseV’s causal ML model predicted weight change, side effects, and dropout in GLP-1 trials using early patient data, but independent clinical review is still needed.
Read more →
SimBioSys gets FDA clearance for AI breast-cancer recurrence tool
The FDA cleared an AI tool that estimates breast cancer recurrence risk using imaging, pathology, and clinical data.
Read more →
Latest AI News for Hospitality and Events
Shanghai Electric presents industrial AI and robotics at CIIF 2026
Shanghai Electric showed industrial robots, cooling tech, and AI software for energy management and predicting equipment repairs at a 2026 trade fair; all performance claims are vendor-reported.
Read more →
Latest AI News for Human Resources
TriNet adds AI and embedded HR tools for small businesses
TriNet is adding AI-powered HR tools for small businesses to automate recruiting, leave, spending, and analysis, but warns companies must assign clear compliance and data oversight roles.
Read more →
AI surveillance at work expands as firms track employees minute-by-minute
AI surveillance now grades workers minute-by-minute through keystrokes and mouse movements, with teachers at Euan Blair's firm reporting "horrendous stress" from automated scoring.
Read more →
OpenAI agrees not to use Apple trade secrets during lawsuit
A court order now stops OpenAI from using Apple trade secrets in their legal fight and requires device images to track what data was accessed.
Read more →
Fintech CMOs plan bigger teams as AI changes marketing work
Fintech marketing leaders plan to expand teams and budgets while using AI mainly for routine tasks, keeping humans in control of sensitive communications.
Read more →
SixFifty launches AI agent for HR compliance
SixFifty launched an AI tool that tracks employment law changes and helps HR teams update policies and paperwork, so staff can focus on higher-risk legal work.
Read more →
Latest AI News for Insurance
Latest AI News for IT and Development
Advantech and DEKRA partner on industrial cybersecurity compliance
Advantech and DEKRA are teaming up to combine industrial computers with certification support, aiming to help manufacturers meet cybersecurity rules during development.
Read more →
Perplexity open-sources SPACE agent sandbox
Perplexity open-sourced its SPACE agent sandbox, giving developers a secure, isolated environment to run and recover agent tasks without exposing credentials.
Read more →
GitHub launches AI Scan for pull requests to find vulnerabilities CodeQL misses
GitHub's AI Scan for pull requests is now generally available, covering PHP, Shell, Dockerfiles, Terraform, and other languages that CodeQL does not support. Findings are advisory only, cannot block merges, and consume Code Security AI Credits without requiring a Copilot license.
Read more →
Nvidia releases Nemotron 3.5 Super VL and two Bolt models for document analysis and local agent tasks
NVIDIA released Nemotron 3.5 Super VL, a 120-billion-parameter open model combining language and vision that enterprises can post-train on proprietary documents. After customization, medical-claims accuracy jumped from roughly 28% to 72%, outpacing the Claude Opus 5 score of about 38% on that task.
Read more →
Latest AI News for Legal
Microsoft launches Copilot for Legal with specialized skills
Microsoft launched a legal AI tool with skills for drafting and contract review, but warns users must verify its outputs.
Read more →
Fort Hill opens AI construction payment audits to project owners
Fort Hill launched an AI tool that checks construction payment requests against contracts, flags issues, and still requires an auditor’s approval before payment.
Read more →
