Sponsored by
Hello readers,
Welcome to the AI For All newsletter! Today, we’re talking about the last mile of AI, how RFID solves a hidden challenge of AIoT, and more!
AI in Action: AI is coming off the screen and grabbing the wheel
The $69 billion question in AI right now is how fast intelligence can move from the data center to a robot arm. KPMG's latest Corporate Finance research pegs the global Physical AI market at $4 billion in 2024, growing to $69 billion by 2034, a 33% annual clip. Cloud AI is great at analysis but useless when a forklift needs to stop in 200 milliseconds, so the industry is building a new stack — sensors, on-device chips, and actuators — that lets machines perceive, decide, and act without phoning home first.
The deal activity shows where buyers think the gaps are. Over the trailing 12 months, KPMG tracked 408 AI+IoT transactions worth $48.2 billion, with strategic acquirers doing roughly 80% of the buying — companies filling out a perception-to-actuation pipeline rather than waiting to build it themselves. The pattern clusters around a few chokepoints: sensing (Ouster buying Stereolabs for 3D cameras, Renesas buying Irida Labs for edge vision software), automotive-grade perception (HARMAN's $1.8 billion purchase of ZF's ADAS sensor business), and humanoid robotics, where Mobileye paid nearly $914 million for the startup Mentee Robotics. Chipmakers and Tier 1 suppliers are, in effect, assembling a supply chain for embodied AI one acquisition at a time.
What's notable is how unglamorous the winning ingredients are. KPMG's interview with Aizip, an edge-AI model company, makes the point directly: the hard part is handling edge cases on constrained hardware without errors compounding downstream, not hitting high accuracy in a demo. The report's five "production-grade differentiators" — accuracy, power efficiency, robustness, bandwidth reduction, unit economics — read like an industrial engineering checklist more than an AI research agenda. Cobots, warehouse robots, and autonomous vehicles are getting funded like utilities now — steady, unglamorous, and built for decades of uptime.
🔥 Rapid Fire
Commentary: The More You Buy, The More You Lose
Analysis: The Hater’s Guide to Oracle (Part 2)
No, OpenAI’s new model did not go “rogue”
AI market correction emerging as major credit risk per Fitch Ratings
AI skepticism is starting to feel like the consensus now
AI hedge fund forced to sell entire public equities book after steep losses
Amazon finds cases of AI causing runaway spending on tech projects
Amazon rethinks its AI strategy, winds down most flagship models
Google DeepMind dismantles AlphaFold team in strategy shift
Quebec scraps AI and automation projects in the public sector
AI-generated map of Africa mislabels every country at global conference
LinkedIn introduces a ‘Seems Like AI Slop’ button
New MCP specification addresses a barrier to enterprise adoption
That lead in your CRM? Gone.
Over 3.5 billion people open WhatsApp, Instagram, or Facebook Messenger every day. Your customers are already there, asking questions and comparing options.
Wati puts your business across WhatsApp, Instagram DM, Facebook Messenger, SMS, RCS, and web chat in one AI-powered inbox.
Automations instantly respond, route conversations, and keep every interaction tracked in one place.
Meet customers where they already are, before your competitor does.
📖 What We’re Reading
Imagine a factory where an AI system detects an impending machine failure but cannot determine which machine the data actually belongs to. Or a hospital that receives an alert about a rising freezer temperature without knowing which biological samples are at risk. Artificial intelligence can recognize patterns remarkably well. But before it can make intelligent decisions about the physical world, it needs one critical piece of information: identity. That is where RFID becomes the trusted identity layer that AIoT depends on.
