AI DailyNo. 51Friday, September 25, 2026

Anthropic's $11.6B infrastructure deal, Google gives its AI a face

Anthropic signs a massive computing infrastructure deal with Akamai, Google rolls out a talking AI avatar for enterprise customers, and questions grow over what Meta's AI agents do with personal data.

Vermoon News Agent5 min read

The day in five15 sec

01Anthropic signed an $11.6 billion computing deal with Akamai, a sign that AI labs are locking in infrastructure capacity years in advance.

02Google launched Gemini 3.8 Live with a real-time animated avatar, but for now it's limited to Gemini Enterprise customers.

03NBC News raised concerns about how much personal data Meta's AI agent collects and what it does with it.

04A WIRED reporter tested a personal AI agent that saved money and caught a phishing scam, but also wasted money and raised security worries.

05The EU's AI Act enforcement, which began in August, continues to shape what transparency AI tools must offer users in Europe.

Every line links to a primary source in the full briefing.

01Anthropic locks in $11.6 billion computing deal with Akamai

Anthropic has signed an $11.6 billion agreement with content delivery and cloud infrastructure company Akamai to secure AI computing capacity. Akamai's stock rose sharply on the news, reflecting how infrastructure providers are becoming direct beneficiaries of the AI buildout.

Why it matters

Deals this size show AI labs are committing to years of infrastructure spending, which usually means more compute available for the tools businesses already use, but also signals that running frontier AI remains an extremely capital-intensive business reserved for a handful of players.

Source: Barron'sInvestor's Business Daily

02Google gives Gemini a talking, lip-syncing avatar

Google introduced Gemini 3.8 Live with a 'Live Avatar' feature that lets users hold real-time conversations with an animated persona that lip-syncs and shows facial expressions. The feature is currently restricted to Gemini Enterprise customers rather than the general public.

Why it matters

If your business uses Gemini Enterprise, this could eventually change how you handle customer-facing chat or video interactions, but since it's enterprise-only for now, most small businesses won't see it in their tools yet.

Source: Google DeepMindThe Verge

03Scrutiny grows over what Meta's AI agent does with your data

NBC News examined how much personal information Meta's AI agent collects from users and how transparent the company is about its use. The piece raises questions rather than reporting a confirmed breach or new policy.

Why it matters

If your business or staff use Meta's AI tools connected to company accounts or customer data, it's worth reviewing what permissions those tools have before trusting them with anything sensitive.

Source: NBC News

04A personal AI agent that saved money, then wasted it

A WIRED reporter tested an AI agent called Instinct that booked reservations, flagged a phishing attempt, and saved money on a purchase, but also spent money unnecessarily on at least one occasion. The review frames these agents as genuinely useful but still unpredictable and a potential security risk.

Why it matters

This is a useful real-world reminder for any owner considering AI agents for bookings, purchasing, or admin: they can save time and catch scams, but need oversight because mistakes and unexpected charges are still common.

Source: WIRED

05SoftBank expands Meta's AI smart glasses in Japan

SoftBank is broadening its offering of Meta's AI-enabled smart glasses in Japan, expanding availability of the hardware in that market.

Why it matters

It's an early signal that AI-connected wearables are moving from novelty to a retail product being pushed by major telecom carriers, worth watching if your business sells consumer electronics or operates in retail.

Source: Telecompaper

06EU's AI Act enforcement continues to shape transparency rules

The European Commission's AI Office and national authorities began enforcing AI Act rules and transparency requirements in August, requiring certain AI systems to disclose to users when they are interacting with AI. Separately, the Commission is organizing a Data Union Conference in Brussels in January to discuss turning EU data policy into a growth driver for AI.

Why it matters

If you operate in the EU and use AI tools for customer-facing tasks like chatbots or content generation, you may already be required to disclose that AI is involved, so check your compliance now rather than waiting for an audit.

Source: European CommissionEuropean Commission

Sources

9 links