February 13th-February 19th // Estimated Reading Time: 8 minutes

Apple and Meta race to put AI on your body
Source: TechCrunch, The Information
February 18-19, 2026

Summary: Apple is accelerating development of three AI-powered wearables. They include AI-enhanced AirPods, an AI pin that can be clipped to your shirt, and smart glasses with a high-resolution camera that could be released in 2027. All three will lean heavily on Siri and connect to the iPhone (and remember, Apple will be partnering with Google to have Gemini power Siri). Meanwhile, Meta is reviving a smartwatch project🔒it killed in 2022. Now code-named Malibu 2, it has a planned 2026 launch featuring health tracking and a built-in Meta AI assistant.

Opinion: With its $799 Ray-Ban Display sold out internationally, Meta may have an early edge in the AI-powered smart glasses race. But here comes Apple! When (not if) AI-powered wearables take off, they will generate loads of rich, unique behavioral data that could fuel better advertising. Meta is an advertising company first and foremost; it will almost certainly use smartwatch health and activity data to improve ad targeting and measurement (imagine if Meta reports on average user heart rate for campaigns and offers “cost-per-heartbeat” pricing 🤪🤯). On the other hand, Apple has built its brand on being “anti-advertising” and data privacy-centric, so who knows how ads will factor into its AI wearables monetization strategy, if at all.

Anthropic's newest model, Sonnet 4.6, does more for less

What It Does: Anthropic just upgraded its default model for Claude. Sonnet 4.6 is faster and cheaper than Anthropic's premium "Opus" tier, and now matches or beats it on many everyday tasks like coding, document analysis, and navigating software like a human. We even used it to edit this newsletter. How’d it do?

Quick Take: Anthropic's enterprise business has exploded, with customers spending $1M+ annually growing from 12 to over 500 in two years. Making powerful AI free could go a long way toward helping Anthropic build more of a consumer-facing AI business (they clearly want to, or they wouldn’t have run a ChatGPT attack ad during the Superbowl).

Google's Pomelli Photoshoot can turn product photos into studio-quality marketing assets

What It Does: Google Labs' free tool Pomelli Photoshoot lets businesses upload product photos and generate professional studio or lifestyle shots in a few clicks, powered by Google's Nano Banana image generation model. 

Quick Take: Professional product photography can be a barrier for small businesses competing against larger brands. Google is offering this for free to lock small businesses deeper into its advertising ecosystem. Here’s the pitch: “Create a great ad using Pomelli, use it to launch a Performance Max campaign, and you’re off to the races!” Google is using AI as a loss leader as much as possible to get customers even more hooked into its existing paid products. Google is good at this.

The Setup: HubSpot, a customer relationship management (CRM) platform, had an ironic internal problem: its own teams couldn't give customers the attention they deserved. Customer success managers spent too much time crafting personalized content for top accounts. Engineers were buried in migrations and maintenance instead of shipping product. Marketers were constantly rebuilding campaign context from scratch.

The AI Solution: HubSpot deployed Claude for all three teams. Marketing built shared Claude Projects with brand guidelines and campaign context baked in. Engineers connected Claude Code directly to HubSpot's internal systems, which allowed them to compress months of migration work and onboard new engineers faster. Customer success managers used Claude to prep for every customer call and generate personalized follow-ups across their entire book of business.

The Results:

Why This Matters: Each of these wins is meaningful on its own. But there's a compounding argument to be made for deploying the same AI across all three functions. When marketing, engineering, and customer success all work inside the same AI system, institutional knowledge stops living in silos. The brand voice that marketing encodes into Claude Projects can inform how CS communicates. The customer patterns that CS surfaces can feed back into what marketing prioritizes. The product context that engineering builds can make CS conversations smarter. A unified AI layer is essentially a shared brain that gets more useful the more of your organization feeds into it.

Your Action: Before deploying AI team by team, ask whether the tools you're evaluating can share context across functions or whether you're building more silos. The productivity wins from any single AI deployment are real but bounded. The larger opportunity is in tying together your marketing context, your customer data, and your product knowledge, all within the same system, in a way that any team can draw from all of it. 

Ad tech vets raise $15M to build AI agents for marketers - The founders behind Rapt, Krux, and super{set} have launched Kana to build AI agents that can drop into legacy tech stacks and help marketers with targeting, analytics, and more. 

AI isn't cutting marketing jobs, but it is making them harder🔒- A survey of 501 advertising decision-makers found that 67% report little to no AI-driven layoffs, but nearly two-thirds say their daily responsibilities have already shifted toward faster output and more complex work.

CMOs are adopting AI more slowly than it's evolving🔒- Major brands still have humans anchoring 85% of their AI workflows because governance, accountability, and fragmented data make full autonomy a liability question as much as a technology one.

Unilever signs 5-year Google Cloud deal to prepare for agentic commerce - The CPG giant will migrate its data to Google Cloud and develop enterprise-wide AI infrastructure to ensure its brands get discovered as more consumers shop through AI assistants rather than traditional search.

Airbnb says AI chatbot traffic converts better than Google - CEO Brian Chesky revealed that users coming to Airbnb via AI assistants like ChatGPT and Gemini are more likely to book than those coming from Google search, suggesting AI search sends fewer, more purchase-ready customers.

What Q4 earnings calls tell us about AI search - Shopify said AI-driven orders are up 15x since January 2025, and Google is seeing AI Mode queries run 3x longer than traditional searches. 

Hightouch launches Content Assembly to unclog creative production - The new feature assembles on-brand campaign assets from a marketer's existing creative library instead of generating new content from scratch. 

“AI doesn’t have a point of view.” Tom Riordan, U of Digital AI expert and founder of AI consultancy WordPower, writes about where humans can make their mark in the AI era.

OpenClaw creator Peter Steinberger is now joining OpenAI - It was only a few weeks ago that Steinberger's open-source AI project went viral, letting users easily create personalized AI agents that can handle tasks like making restaurant reservations. Now OpenClaw (and it’s one employee) has been snapped up by OpenAI for a rumored $1B+!

The U of Digital AI Edge Newsletter is intended for AI Edge Pro subscribers, but occasional forwarding is okay!

To subscribe visit Uof.Digital/Newsletters or AI Edge Pro, or contact us directly for group subscriptions.

And remember, U of Digital helps teams drive better outcomes through structured education on critical topics in marketing / advertising like AI, ad tech, mar tech, privacy / identity, CTV, commerce media, and more. Interested in learning more about how we can supercharge your team? Let’s talk!