Today's AI Pulse
Salesforce to acquire AI‑agent platform Fin for $3.6B
Salesforce announced a $3.6 billion purchase of Fin, the AI‑agent platform formerly known as Intercom, to strengthen its Agentforce suite. The deal comes as Salesforce reported strong Q1 results and a $50 billion buyback program. Fin’s technology enables automated customer support across chat, email, WhatsApp, phone and Slack.
Also developing:
By the numbers: Waymo acquires Apple’s Arizona Proving Ground for $220M
The Fastest Way to Build an AI Agent (Start With the Output, Not the Tool)
The article proposes a reverse‑engineering method for building AI agents by starting with a mock output rather than tool capabilities. By drafting a fake example of the desired result, users can instantly identify required inputs, define the schema, and streamline integration. This approach, illustrated with meeting‑summary automation, reduces ambiguity and accelerates development. The author recommends a 10‑minute exercise to validate readiness before building.
SF Offers AI Events; My AI Curates Them
One thing San Francisco has that even San Jose doesn't is tons of events for AI people and @michelleefang keeps track of them. So does my AI at https://t.co/8L5xphk0qQ, mostly by reading Michelle's posts here. In fact, I talk to...
NightBeacon Slashes Detection to Seconds, Cuts False Positives 87%
NightBeacon is changing the game for us @Binary_Defense - we have this technology so dialed into every aspect of what we do. Our MTTD and MTTR has drastically been reduces to seconds and minutes, not multiple minutes or hours....
Pinterest Recommendation Engine Update Aims For Enhanced Ad Relevance
Pinterest unveiled a Contextual Sequential Two‑Tower Model that injects real‑time browsing signals into its ad candidate generator, delivering a 3‑10× lift in per‑user ad relevance and boosting median relevance by roughly 275‑300%. The upgrade addresses a prior shortfall where less...
Microsoft Researchers Find AI Models and Agents Can't Handle Long-Running Tasks
Microsoft researchers published a preprint revealing that even top‑tier large language models (LLMs) corrupt documents during long‑running, multi‑step workflows. Using a new benchmark, DELEGATE‑52, they measured a 25% average loss of content after 20 delegated interactions, with overall degradation around...
Lendi Group Runs First Project Through "Agentic SDLC"
Lendi Group, a home‑lending fintech, completed its first project using an agentic software development lifecycle (SDLC) powered by Atlassian’s Teamwork Graph. The new process captures meeting audio with Loom, auto‑generates PRDs in Confluence, and employs AI agents to create Jira...
Why 62% of Organizations Still Cannot Scale AI Safely
A Stanford AI Index report shows 62% of organizations view security and risk as the primary obstacle to scaling agentic AI, outpacing technical or regulatory concerns. The same data reveals AI‑related incidents jumped from 30% to 50% of firms between...
Hong Kong Highlights Innovation Partnerships and AI Development During Sweden Visit
Hong Kong’s Innovation, Technology and Industry Secretary Prof. Sun Dong announced a push to deepen global research ties, especially with Sweden, as the city expands its AI and life‑science capabilities. The InnoHK programme now partners with over 30 institutions across...
Mom Community Touts AI Chatbot With A Dig At AI Powerhouses
Peanut, a seven‑year‑old community app for mothers, introduced an AI‑powered chatbot called Ask Peanut alongside an anti‑AI Instagram campaign that lampoons ChatGPT, Claude and Gemini. The campaign stresses that AI lacks the lived experience of pregnancy, sleep regressions and 3 a.m....
‘Deskilling’ Is Bad. This Is Worse.
Harvard education scholars Stephanie Smith Budhai and Marie Heath warn that AI tools in K‑12 classrooms are fostering "never‑skilling," where students never develop foundational abilities like crafting topic sentences. Their research shows generative AI can embed racial and socioeconomic biases,...
RingCentral Boosts 2026 Outlook as AI Customer Base Doubles
RingCentral announced a raised 2026 revenue outlook after the share of paying AI product users more than doubled year over year, now representing over 10% of its customers. The company says AI‑enabled accounts generate higher average revenue per user and...
China's DeepSeek Becomes State‑run AI Powerhouse
Deepseek just casually closed the biggest AI round in Chinese history but most people in the West still don't get it: This isn't a normal startup. The Chinese government is essentially building DeepSeek into the AI company that runs the...
Cloudflare Cuts 1,100 Jobs as AI Adoption Triggers Workforce Reset
Cloudflare announced the termination of more than 1,100 employees, roughly 20% of its global staff, after internal AI usage jumped 600% in three months. The move follows a record $639.8 million first‑quarter revenue and signals a new management model built around...
Starbucks India Appoints Returning Executive as CTO to Accelerate Digital Strategy
Starbucks India announced today that a returning technology executive, fresh from a three‑year stint leading Tim Hortons’ regional IT in Singapore, will serve as its new Chief Technology Officer. The hire underscores the coffee chain’s push to deepen AI‑driven personalization,...
KPMG Survey Shows 93% of US Firms Set to Scale Finance AI, Spotting Talent and Trust Gaps
KPMG released a new survey indicating that 93% of U.S. companies will deploy or scale AI in finance over the next 18 months, with half planning multi‑agent systems. The report flags acute talent shortages, unclear use cases and a growing...
Oregon Judges Fine Lawyers $110K Over AI-Generated Legal Fabrications
A federal judge in Oregon imposed a $110,000 penalty on two attorneys after AI‑driven filings contained invented cases and citations. The state bar reports five Oregon filings and roughly 900 nationwide with similar hallucinations, underscoring a growing risk as lawyers...
Singapore’s SkillsFuture Institute Launches AI‑VR System to Assess and Bridge Worker Skill Gaps
Singapore’s SkillsFuture Institute (SIT) has rolled out an AI‑driven, virtual‑reality assessment platform that tests both technical and soft skills, starting with 50 volunteers at a senior‑care agency and expanding to pilot programs abroad. The system promises faster, scalable evaluation while...
Florida Enacts First AI Data‑Center Energy Law to Shield Ratepayers
Governor Ron DeSantis signed SB 484 into law, requiring large AI‑focused data centers to bear the full cost of electricity and related grid upgrades, preventing Florida utility customers from subsidizing these high‑energy facilities. The measure also gives local governments new...
Ugreen's iDX6011 Pro NAS Brings On-Premise AI to Home Offices
Ugreen unveiled the iDX6011 Pro, a 6‑bay network‑attached storage device powered by an Intel Core Ultra 7 255H and 64 GB of LPDDR5x memory, capable of running local large‑language models. The machine blends enterprise‑grade compute with consumer‑friendly storage, offering dual Thunderbolt 4, dual...
Denies “AI Just Regurgitates” Claim, Cites Hallucination Warnings
Dear @geoffreyhinton, I literally never said that AI systems “JUST regurgitate”; that’s plainly false. I don’t believe it, and I didn’t say it. (They do *sometimes* regurgitate, and the evidence for that is overwhelming.) I further discuss the rest of your...
Agoda Deploys AI to Pair Hotel Photos with Guest Reviews, Boosting Booking Transparency
Agoda has rolled out an artificial‑intelligence system that automatically pairs hotel images with matching guest review excerpts. The tool, first launched on city hotels in Asia, lets travelers see feedback tied to specific visual elements, addressing long‑standing concerns over misleading...
Booz Allen Hires Former Army CIO Leonel Garciga to Drive Defense AI Strategy
Booz Allen Hamilton appointed former U.S. Army chief information officer Leonel Garciga as a senior executive adviser to accelerate defense technology delivery. Garciga warned that rapid AI adoption in the Army is creating a human‑centric tension, underscoring the need for...
China Suspends New Autonomous‑Vehicle Licenses After Wuhan Pilot Chaos
China's transport regulator announced a freeze on new autonomous‑vehicle licenses following a mass halt of Baidu’s robotaxi fleet in Wuhan. The move underscores safety concerns and the need for resilient infrastructure as the industry scales.
Roblox's AI Push for Photorealism Meets Developer Resistance
Roblox announced plans to integrate AI tools that can render games in photorealistic style, but many independent developers on the platform are skeptical, saying the aesthetic shift could alienate the core audience. The clash highlights a broader tension between platform‑level...
In Wake of Spotify “Verified,” Artist Stream Counts Plummet … Sometimes
Spotify launched a "Verified" badge this month to label human‑created music and separate it from AI‑generated tracks. The authenticity criteria require at least 10,000 monthly listeners, 1,000 followers, and verified social signals such as concert dates and merchandise. Shortly after...
ProcurePro Secures $11 Million US Funding to Accelerate AI in $13 Trillion Construction Procurement
ProcurePro, the Brisbane‑based construction procurement platform, closed a $11 million US (AU$15 million) round led by QIC Ventures with French builder Bouygues joining as a backer. The capital will fund a global AI product roadmap, a new US office and the hiring...
SAP Announces Dual Acquisitions of Dremio and Prior Labs to Bolster Data Lakehouse and AI Governance
SAP announced on May 4, 2026 that it will acquire Dremio, an Apache Iceberg‑native lakehouse, and Prior Labs, a Tabular Foundation Model pioneer. The moves are designed to give SAP a unified data foundation and tighter AI governance, positioning the company against...
AI Commerce Platform District Secures $14.7M Seed Funding Led by Andreessen Horowitz
District announced a $14.7 million seed round headed by Andreessen Horowitz and Kindred Ventures. The funding will fuel its general‑availability launch, expanding a platform that lets creators and brands build interactive, code‑free storefronts using AI.
Faith and AI Converge: Guiding Moral Future
My dear friend and partner in the Faith-AI Covenant Project, Dana Humaid Al Marzouqi, and I come from different cultures, different professional backgrounds, and different faiths. But we share a deep conviction that humanity is approaching a turning point with...
AI Model Firms Boost Value for Chip and Cloud Giants
We may have AI model companies that don't have a clear pathway to revenue, but that doesn't mean they're not creating value. They are creating value for Nvidia, Intel, Micron, Sandisk, and distributors like the Apples of the world, or...
OpenAI Rolls Out ChatGPT Ads Manager Beta, Opening New Channel for Marketers
OpenAI today launched a beta version of Ads Manager, a self‑serve portal that lets advertisers run campaigns inside ChatGPT. The platform relies on conversational context rather than keyword matching and promises transparent labeling and user‑controlled data use.
Datadog Posts $1B Q1 Revenue, Twilio Unveils AI Agent Tools, Shares Jump 31%
Datadog reported quarterly revenue exceeding $1 billion and lifted its full‑year outlook, sending its shares up 31% on May 7. Twilio announced AI‑agent platform upgrades at its annual conference, positioning the communications SaaS firm for new revenue streams. Both results underscore AI...
Colin Angle Launches “Familiar,” A Lifelike AI Companion Robot
Colin Angle, the co‑founder of iRobot, announced the upcoming launch of “Familiar,” a furry, AI‑driven domestic robot that will sell next year. The device aims to provide genuine companionship through on‑device generative AI, positioning itself as a pet alternative in...
Circle Secures $222 Million From BlackRock, Apollo and A16z for Arc Blockchain
Circle announced a $222 million raise led by Andreessen Horowitz, with participation from BlackRock, Apollo and others, valuing its Arc token at $3 billion. The funding backs a public blockchain aimed at institutional finance and AI‑driven applications, expanding Circle beyond its USDC...
Why AI Funding Is So Price-Insensitive
In this brief episode, Morgan Stanley’s Global Head of Fixed Income Research, Andrew Sheets, explains why AI‑related capital spending is remarkably price‑insensitive. He highlights the massive $800 billion U.S. tech investment this year—nearly double last year’s spend—and notes that component costs...
AI Ethical Minefields in Clinical Decision-Making
Hospice providers are testing AI for documentation, operational efficiency and clinical decision support, aiming to offset rising demand and shrinking staff. Dr. Kimberly Curseen of AAHPM argues that smart AI can lift administrative burdens without eroding patient rapport, but regulatory...
HPE Memory Server Targets Compute-Heavy and Agentic AI Workloads
HPE announced the Compute Scale‑up Server 3250, a memory‑centric platform built on the Superdome Flex architecture and powered by Intel Xeon 6 processors. The system scales from four to 16 sockets and can hold up to 64 TB of DDR5 RAM, targeting compute‑heavy workloads,...
SessionStart Hook Cuts Token Usage by Half
The nav-start skill used to make 6 Read calls at session start. Replaced them with a Claude Code SessionStart hook that injects state via additionalContext — before the first user turn. Same data, zero reads, ~35k tokens saved per session 🔥 Measured: 73.3k →...
AI Floods Code, Outpaces Trust and Triage
"AI increased the supply of code faster than it allowed maintainers to triage contributions. AI didn’t increase the supply of trust yet. That is the real open-source problem: AI is not killing open source. It is killing open contribution as we...
Sigma Browser Adds Hermes Agents for Local AI Workflows
Sigma browser now supports Hermes Agents 🤯 You can now run open-source models natively via their Eclipse engine for unlimited agentic workflows. Private. Local. Open-Source. No cloud. No subs. Mac. Windows. iOS. Android 👀↓ https://t.co/tMwwPolCq9
Weekend Roundup: Top AI Marketing News Highlights
ICYMI Top #Marketing #AI News Stories from over the weekend: https://t.co/8V6t1LWrNB Top stories from @FastCompany @InsideIowaState @FiercePharma @martechismktg @CNBC @BainandCompany @saastr and @Oracle Blogs. https://t.co/21uxv4SV69
Exploring Agentic Harness Use and AI Bubble Debate
I am curious on people's responses - have you: - used an agentic harness - and whether you think AI is a bubble
AI Transforms Every Stage of the Buyer Journey
How #AI Is Redefining The Buyer Journey by @ingliguori #Marketing #ArtificialIntelligence #MachineLearning #ML #MI https://t.co/6NjaJb1Nk3
Multilingual, Multitask AI Demo Proves Its High-Value Appeal
The demo of this model is cool. Interacts with multiple people in multiple languages. On multiple tasks. Now we know why @miramurati got the big bucks.
Object Storage Powers AI with Unprecedented Speed
Object Storage is having a moment. This boring, reliable part of a cloud's portfolio matters more now for AI training and inference. @googlecloud Storage Rapid is amazing. 20m queries per second? 15+ TB/s read throughput? Bonkers numbers. https://t.co/d3fWXnuy1k https://t.co/eaemRlmjjD
ChatGPT Generates $39K, Just 0.2% of Sales
Agentic Commerce on a real store: $39,015 total sales attributed to ChatGPT or 0.2% of store revenue for the same period. Yes, AI makes sales... but it's also still a rounding error. Both can be true. https://t.co/xubN6B0T4r
Quantum Adapters Cut Llama 3.1 Perplexity 1.4%
Quantum circuits are boosting Llama 3.1 LLM performance. Llama 3.1 with Cayley Unitary Adapters improves WikiText perplexity by 1.4% with only 6,000 additional parameters, running end-to-end on 156 qubit Heron r2. An early, tangible quantum assist for frontier...
Chinese AI Engineers Become Silicon Valley’s New Power Players
Chinese AI engineers are Silicon Valley’s new power players Chinese-born tech workers have fueled Silicon Valley for decades. In the AI era, they’re superstars. https://t.co/SYZkeTSFjf
Policy Is as Crucial as Tech for AI Education
#AI In Education: Why Policy Matters As Much As #Technology by @geoffreyalef1 @Forbes Learn more: https://t.co/2lTtAP3bPg #EduTech #ArtificialIntelligence #DigitalTransformation https://t.co/yKN8LEzvHf
Future Models Will Resolve Past Technical Debt
i’m betting on the bitter lesson. that each new generation of model will fix the technical debt incurred by the the models 2-3 generations ago. slop away.