13 May 2026: Google Unveils AI Laptops and America Creates Its First AI Healthcare Payment (AM)

Google’s Android Show on Tuesday unveiled a new laptop category built around Gemini Intelligence, the US government quietly created the first payment mechanism for AI-powered healthcare agents, and Anthropic stepped into the legal services market. On a day dominated by platform moves rather than model launches, the signal is clear: AI is migrating from standalone tools into the operating systems and professional workflows that people already use.

Google has unveiled Googlebook, a new laptop category designed from the ground up for Gemini AI, and the mouse cursor may never quite be the same again. At the Android Show on Tuesday, Google revealed Googlebooks in partnership with Acer, Asus, Dell, HP, and Lenovo. Unlike Chromebooks, Googlebooks merge Android and ChromeOS into a single platform and put Gemini Intelligence at the centre of the experience rather than treating it as a bolt-on feature.

The headline capability is Magic Pointer, built with Google DeepMind. Rather than switching to a separate AI panel, the cursor itself becomes the entry point. Point at a date in an email and Gemini offers to schedule it. Select two images, say a new sofa and your living room, and the AI visualises them together. Google is also bringing vibe-coded widgets to Googlebooks and Android phones: a generative feature that lets users place custom information panels on their home screen using plain-language descriptions, with no coding required. Googlebooks arrive this autumn. If you are working out how to phase AI into your team’s day-to-day, Cristoniq’s guide to which tasks you should give to AI first is a good starting point before your next device refresh.

The United States has created, for the first time, a payment mechanism for AI agents operating in healthcare. A TechCrunch investigation published early this morning revealed that a new Medicare payment model called ACCESS, introduced by the Centers for Medicare and Medicaid Services, creates a reimbursement pathway for AI-driven patient coordination that happens between clinic visits.

Until now, there was no governmental mechanism to pay for an AI agent that monitors a patient after discharge, calls to check on medication adherence, or coordinates a care referral. ACCESS changes that. It funds the coordination layer that human capacity alone cannot provide at scale, targeting patients with complex chronic conditions who typically fall through the gaps between appointments.

The UK angle matters. NHS England has been piloting AI monitoring tools in primary care, but reimbursement models for between-visit AI activity remain largely unresolved. The US creating a working payment template gives NHS commissioners a framework to adapt. Watch for NHS England or the Department for Health and Social Care to reference ACCESS in forthcoming primary care AI guidance as a signal this approach is gaining institutional traction.

Anthropic is entering the legal AI market, bringing Claude’s document-handling strengths to law firms that want to automate clerical work without replacing their lawyers. A TechCrunch report from Tuesday detailed new Claude-powered tools covering document search and review, case law retrieval, deposition preparation, and document drafting, offered as integrations that legal teams can embed in existing workflows.

The legal AI space has been building for over a year, with Harvey and Clio among the established names. Anthropic’s entry matters because Claude handles very long documents and complex multi-step instructions with notable consistency, both of which matter in legal work where errors are costly. For small law firms and solicitors in the UK, this signals that enterprise-level legal AI is moving closer to their market. For context on how Claude differs from older chatbot tools, Cristoniq’s explainer on the difference between a chatbot, copilot and agent sets out the practical distinctions.

Here is everything else worth knowing from this morning’s AI news.

• OpenAI Codex is being used by finance teams to build monthly business reviews, variance bridges, and model checks from real inputs, according to new guidance from OpenAI Academy. [12 May]
• Google adds Gemini-powered dictation to Gboard, starting with Samsung Galaxy and Pixel devices, putting pressure on standalone dictation apps. [12 May]
• Google and SpaceX are reportedly in talks to build AI data centres in orbit, pitching satellite compute as future AI infrastructure, though costs remain far higher than ground-based alternatives. [12 May]
• Threads is testing a Meta AI feature giving users real-time context on trends within conversations, working similarly to Grok on X. [12 May]
• Elon Musk reportedly considered passing control of OpenAI to his children, Sam Altman testified in the ongoing legal dispute. [12 May]
• Anthropic has warned investors that secondary market sales of its shares are void and will not be recognised by the company. [12 May]
• Needle, a 26 million parameter tool-calling model from Cactus, is now open source, running at 6,000 tokens per second on consumer hardware. [13 May]
• Voker (YC S24) has launched a public agent analytics platform giving product teams session replays and failure tracking for AI agents in production. [13 May]

The immediate question from today’s news is whether the US ACCESS healthcare payment framework begins to travel. The moment NHS commissioners reference it in procurement guidance, UK health tech companies building between-visit AI tools will find the funding conversation changes quickly. That development is worth monitoring closely over the next few weeks.

This is a daily news update for informational purposes only. AI products and policies change rapidly. Verify details directly with providers before making decisions. Nothing here is financial or legal advice.

AI Daily is Cristoniq’s daily guide to developments in artificial intelligence, published every morning.