Today's AI Pulse
Anthropic shuts down Claude Fable 5 and Mythos 5 after US export‑control warning
Anthropic disabled access to its Claude Fable 5 and Claude Mythos 5 models after the Trump administration, via the Commerce Department, warned that foreign nationals could exploit the systems. The move followed alerts from Amazon and other firms about a potential jailbreak that could aid cyber‑security attacks. Anthropic declined the administration’s request for a voluntary shutdown but complied with the order.
Also developing:
By the numbers: AMD acquires AI memory‑optimization pioneer Mext
Bipartisan Senate Bill Takes Aim at AI Voice Cloning and Deepfake Fraud Targeting Mobile Users
The bipartisan AI Fraud Accountability Act (S.3982), introduced by Senators Lisa Blunt Rochester and Tim Sheehy, would make it a federal crime to use AI‑generated voice clones, synthetic video, or other deepfakes for financial fraud and identity theft on mobile devices. The bill grants the Federal Trade Commission explicit enforcement authority and tasks NIST with creating technical standards and detection benchmarks for synthetic media. It expands on the earlier AI Scam Prevention Act by adding criminal penalties and a dedicated NIST research mandate. The legislation arrives as banks and identity‑verification firms roll out anti‑deepfake tools for mobile authentication.
AI Risks With the Patent Office: What Life Science Companies Should Be Asking Now
Life‑science companies are increasingly using generative AI to draft patent applications, but statements filed with the USPTO become a permanent record. Errors or “hallucinations” produced by AI lack warning labels, exposing firms to potential inequitable‑conduct claims. While AI can boost...
HIMSSCast: Ambient AI Scribes Pose Important Regulatory and Legal Questions
Ambient AI scribes promise to slash clinicians' charting time, driving rapid adoption across hospitals. Yet they are far from plug‑and‑play, requiring careful integration, data governance, and workflow redesign. Legal experts warn that these tools raise HIPAA, privacy, and liability concerns...
Publishers Sue Meta for Pirated Books Used in AI Training
Major publishers sued Meta for pirating millions of books to train its AI Five major publishers are suing Meta for training its AI on pirated books. The lawsuit alleges Zuckerberg “personally authorized” massive copyright infringement. It claims Meta’s Llama AI model was trained...
They Can Read Your Thoughts. Who Will Protect Your Mind?
Neurotechnology now decodes silent thoughts and can subtly steer behavior, turning the human mind into a commercial asset. Researchers have demonstrated fMRI‑based sentence reconstruction and emotion‑AI tools that monitor facial, vocal, and physiological cues. Scholars propose four new human‑rights‑style protections—cognitive...
![[Video] AI Today in 5: May 5, 2026, The Affordable AI Edition](/cdn-cgi/image/width=1200,quality=75,format=auto,fit=cover/https://jdsupra-static.s3.amazonaws.com/profile-images/og.2237_4849.jpg)
[Video] AI Today in 5: May 5, 2026, The Affordable AI Edition
The latest episode of AI Today in 5, hosted by compliance expert Tom Fox, spotlights five AI developments reshaping multiple sectors. Nquiry, an AI‑powered regulatory intelligence platform, promises near‑instant compliance answers. Bloomberg highlights China’s low‑cost large‑language models—DeepSeek, Qwen, Moonshot—pressuring U.S. AI firms....
AI Forces Leaders Back Into the Trenches
Lots of possible takes on today's Coinbase announcement, but this passage caught my eye. If this becomes a trend, it will be quite disruptive to tech's managerial layer. It was something I was acutely aware of during my comms career: the...
Sprinklr CX Connect New York 2026
Sprinklr CX Connect 2026 convenes senior marketing, customer experience and service leaders on June 4 in New York’s Convene 237 Park. The half‑day event focuses on AI‑driven innovation, digital ecosystem transformation, and emerging business models reshaping brand‑consumer interactions. Attendees will engage in peer networking,...
AI Is Saving Pharma Billions in Manufacturing and Back-Office Work, Just Not in the Lab
Eli Lilly’s digital chief says AI is delivering cost savings across pharma, but not in drug discovery where expectations were highest. The company’s AI‑driven digital twin of tirzepatide manufacturing cut production time and lifted output, echoing similar back‑office efficiencies at...
Charting Change in Legal: AI-First Law Firms, Big Tech, and the Future of Legal Work
In the latest episode of Charting Change in Legal, analysts Ari Kaplan and Caroline Hill dissect the rise of AI‑first law firms, exploring how venture funding and new business models are redefining the legal services market. They examine big‑tech giants...
Persefoni Launches Agentic AI to Help Companies Analyze Carbon Footprint
Persefoni, a sustainability‑software firm, unveiled the Persefoni Analytics Agent, an agentic AI feature that lets users query emissions data with plain‑language prompts. The tool instantly creates charts, tables, and visual explanations, streamlining carbon‑footprint analysis for corporations and financial institutions. By...
Komprise Patents Dynamic Load Balancing Tech
Komprise has secured US 12566637‑B2 for its Elastic Shares technology, which dynamically subdivides large unstructured data sets across multiple compute engines for faster AI processing. The patented system uses a job‑supervisor to monitor engine status and reassign work instantly, eliminating idle...
Richard Dawkins Concludes AI Is Conscious, Even if It Doesn’t Know It
Renowned evolutionary biologist Richard Dawkins spent three days conversing with Anthropic’s Claude and OpenAI’s ChatGPT, emerging convinced the bots exhibit consciousness despite lacking self‑awareness. He described the AI’s poetic output and philosophical reflections as “human” and urged open‑minded inquiry into...
A Bill in France Would Force AI Firms to Prove They Didn’t Use Copyrighted Content to Train Their Tech. The...
A coalition of 81 French cultural and media organisations is urging the National Assembly to debate the Darcos bill, which creates a rebuttable presumption that AI providers have used copyrighted works unless they can prove otherwise. The French Senate passed...
Publishers Sue Meta Platforms over Alleged AI Training Copyright Infringement
Publishers Elsevier, Cengage, Hachette, Macmillan and McGraw Hill, along with author Scott Turow, have filed a class‑action lawsuit in Manhattan federal court accusing Meta Platforms of illegally using millions of books, journal articles and novels to train its Llama large‑language model. The...
PROPTECH-X : RoRo AI to Level Playing Field for Independent Brokerages in the US
Rodland Real Estate is launching RoRo, an AI‑driven market‑intelligence platform, in the United States as part of a phased global rollout. The voice‑enabled system pulls live MLS and other data, delivering real‑time insights, reasoning and decision guidance directly within agents’...
First, AI Slop. Now, 'AI Beige.'
The article introduces "AI beige," a term for the bland, homogeneous output that generative AI often produces, eroding brand distinctiveness. It cites EY’s "sameness trap" report, which found teams unknowingly creating identical brand concepts when using AI. The piece argues...
AI and ‘the Art of the Possible’: HR Leaders From BC Hydro, CBC Weigh In
HR leaders at BC Hydro and CBC are exploring AI to sharpen strategic workforce planning, close emerging skill gaps, and boost productivity. Both executives stress the need for robust data architecture and data‑literacy training to empower HR business partners. They...
Where Did Your Web Traffic Go? How to Reclaim Leads From AI Answer Engines
In 2026 AI answer engines are delivering zero‑click answers, siphoning traffic from life‑science marketers’ websites. The AI scans public pages in real time, providing scientists with quick answers and leaving the source unvisited. Marketers are urged to adopt a dual‑layer...
Built In Launches the First Employer Intelligence Platform for AI Search
Built In unveiled the Employer Intelligence Platform, the first solution that lets companies monitor and shape how they appear in AI‑driven search tools like ChatGPT, Claude and Google AI Overviews. The platform delivers analytics, content creation and distribution across a...
Project ARIA Pushes AI From Concept to Soldier-Ready Capability
The U.S. Army’s Project ARIA, launched in March, is moving artificial intelligence from experimental labs to soldier‑ready tools. By partnering with industry and academia, the initiative targets high‑value use cases such as logistics, policy compliance, and daily workflows to cut administrative...
OpenGov Expands AI-Native Platform to Unify ERP, HCM, and Public Sector Operations
OpenGov announced a major upgrade to its Public Service Platform, merging its ERP system with human capital management (HCM) on a single data foundation. The AI‑native suite now includes OG Assist, an embedded assistant that guides users through tasks, data...
California Bar Proposes Rule Requiring Lawyers to Verify Every AI Output — and Five Other AI-Focused Ethics Changes
The California State Bar’s professional‑responsibility committee has proposed amendments to six Rules of Professional Conduct that embed specific AI obligations, moving from guidance to enforceable rules. The changes require lawyers to independently verify any AI‑generated output, disclose AI use when...
The AI Upload: Verifone Adapts AI for Fuel and C-Store Merchants; SoundHound AI Debuts an Agentic Platform
Verifone is embedding generative AI into its Commander point‑of‑sale platform, adding an AI‑driven help center, real‑time fraud‑detection scoring and a self‑checkout integration for fuel and convenience‑store merchants. The AI assistant also supports help‑desk agents by transcribing calls, surfacing relevant documentation...
Artificial Intelligence and the Modern Cybersecurity Landscape
Artificial intelligence is reshaping the cybersecurity landscape, offering efficiency gains while introducing novel threats. Rent the Runway’s VP of engineering is rolling out an AI governance and risk‑management framework to safeguard innovation, and TCS warns that AI tools can amplify...
AI Can Discover and Execute Tasks You Never Imagined
Everyone uses AI the same way. That was me until a few weeks ago. You open a window, type what you need, and it gives you something back. That's useful, but it has a ceiling most people don't realize: you can...
Cisco, Schneider Electric Call for Enabling Regulations to Help AI Flourish
At the SelectUSA Investment Summit, Cisco and Schneider Electric urged the U.S. government to adopt cohesive AI regulations, workforce‑training programs, and power‑grid upgrades. They referenced the 2025 AI Action Plan and a December 2025 executive order aimed at eliminating a...
New Bill Would Narrow Scope of Colorado’s Landmark 2024 AI Law
Colorado Democrats introduced Senate Bill 26-189 to sharply narrow the state’s 2024 AI law, SB 24-205, by replacing the broad “high‑risk artificial intelligence systems” definition with a limited “automated decision‑making technology” (ADMT) framework. The proposal eliminates proactive risk assessments and...
Jamie Dimon and Dario Amodei Shared a Stage for the First Time. Here’s What They Revealed About AI, Cyber Risk...
Anthropic unveiled Claude Opus 4.7, its most capable financial AI model, alongside a suite of ten pre‑built agents for back‑office tasks at a private briefing in New York. The launch coincided with a $1.5 billion joint venture backed by Blackstone, Hellman & Friedman,...
Focus on Tasks: Thriving in AI-Augmented Careers
AI is going to change every job on the planet. That can sound alarming, but the reality is more nuanced. In this video, I explain why some jobs will be displaced, why many more will be augmented, and why the...
Industry Success Demands Relocating to Its Geographic Hub
"The single most important thing for anybody wanting to break into any industry is go to the headquarters or cluster of that industry. Move to wherever that thing is. And all the advice that you can do anything from anywhere...
I Built a Complete Sales Funnel in an Afternoon. Here's the Prompt Sequence.
Solopreneurs can now build a full digital‑product sales funnel in a single afternoon using a nine‑prompt AI workflow. The sequence starts with a buyer‑pain diagnosis, then creates opt‑in page copy, a three‑email welcome series, and finishes with sales‑page, objection, checkout,...
The Federal Safety Net Isn’t Ready for Artificial Intelligence
The article warns that the U.S. unemployment insurance system, limited to 26 weeks of benefits, is ill‑suited for potential AI‑driven job displacement. Existing job‑retraining programs are outdated or have lapsed, and recent Republican legislation has tightened eligibility for food and...
AI Co-Scientists Poised for Nobel-Level Discoveries
Academics and companies alike are developing agents that can carry out research tasks autonomously and work with scientists as genuine collaborators. Some believe these AI co-scientists will one day reach Nobel Prize–worthy heights.
AI Coach Reviews Meetings, Aligns With Performance Feedback
"It's really been surprising how good of a coach [AI] is." @rywiggs (Mercury VP) built a Claude Code system that reviews his meeting transcripts from @meetgranola and cross-references them against his performance review and coaching feedback. "It tells me, hey, in...
4 Ways HR Uses AI in Hiring, According to Talent Professionals
Talent acquisition teams are increasingly deploying artificial intelligence, with screening emerging as the top use case, according to a May 2026 iCIMS and Aptitude Research report. The same study shows AI is also applied to candidate communication, assessments and sourcing, but...
Cross‑Device Handoff Untethers AI Coding with TRAE SOLO
TRAE SOLO Mobile is here, and it changes how AI coding works. the killer feature isn't the mobile app itself. it's the cross-device handoff. assign a Cloud task from your phone. open the desktop or web app. the agent is already running, with full context,...
AI Inferencing Nodes Will Soon Blanket the Planet
Quote of the day comes from Sean Farney, VP, Data Center Strategy – Americas at JLL, here at @ConnectX_USA #ConnectX26 “We need to carpet the planet with AI inferencing compute nodes....We’re in the infancy of this.” #EdgeComputingReturns
Not Just Cars: Watch The Robotic Shift At Auto China 2026
Auto China 2026 highlighted a shift from vehicle‑centric AI to "Physical AI," where the same intelligence platform powers cars, humanoid robots and other machines. XPeng unveiled its Iron robot, built on the Vision‑Language‑Action stack used for driver assistance, while Xiaomi...
Agentic AI Enters First Investable Cycle via Measurable Work
Latest report @DiligenceStack From AI Usage to AI Earnings Power Agentic AI’s first investable cycle is forming around measurable work, workflow control, and budget formation. https://t.co/GO5Exh5d9w
Choosing the Ideal AI Tool for Researchers
What does ‘the right AI tool’ for researchers look like? Maxim Khan, SVP for Academic and Government Solutions, Elsevier, answers the question. https://t.co/mmPAqitTxE #AI #Research #Researchers
5 Questions for MSIG USA’s Ron Morrison
Ron Morrison, chief claims officer at MSIG USA, says the insurance‑claims landscape is being reshaped by AI and broader claims transformation. He stresses that change management, trust and thorough training are prerequisites for any AI rollout to succeed. By automating...
AI Will Dismantle Middle Management, Empower Value Creators
AI is enabling companies to replace managers Middle-management bureaucracy is going to be eliminated. Intelligence at the center, AI-powered value-creators at the margins.
NIST Should Conduct Public, Independent AI Capability Tests
In addition to the CAISI evaluation, it would be useful if NIST conducted public tests of AI abilities as an independent evaluator - though those obviously should not be pre-release tests & can be done when models are public. Independent testing...
Patient-Facing Revenue Cycle Starts at the Front-End
Curae is launching a patient‑facing revenue cycle that begins at the front end of care, using artificial intelligence to pull together physician notes, clinical history, and claims data. Matt Fisher, VP of Operations, says the AI creates a single, comprehensive...
VS Code Auto‑adds Copilot Co‑author Tag Even Disabled
VS Code was adding 'Co-authored-by: Copilot' freely, even when the assistant wasn't used and AI chat features were disabled. https://t.co/UbnKRrWdxo
AI Swapped Effort for Time, Not Productivity
AI didn't solve my productivity problem. It replaced the constraint I understood (effort) with one I'd already beaten once (time). https://t.co/VEPUVUMcoe
Connected by Design: How AI and Automation Are Transforming Drug Discovery at BMS
Bristol Myers Squibb (BMS) is shifting from isolated AI tools to an integrated, learning‑driven ecosystem that connects data, models, and automation across discovery and development. The company highlights its shared data backbone, AI co‑scientists, and lab‑in‑the‑loop automation as foundational layers,...
Skepticism Over AI Data Traffic Forecast's Broad Scope
I’m not entirely convinced by this prediction from @WIAorg about AI data traffic going to 10% on wireless, unless it’s a “kitchen sink” figure including any data that is AI created, modified or recommended (eg timelines) https://t.co/6UoCVNyH72
Top Corp Illegally Trains AI on Pirated Authors' Books
“I find it distressing and infuriating that one of the top-10 richest corporations in the world knowingly used pirated copies of my books, and thousands of other authors, to train Llama, which can and has produced competing material,” Turow https://t.co/wCFSd4NIR0...