Executive Summary

The EU Commission accuses Meta of disadvantaging competitors in accessing WhatsApp data and thereby favoring its own Meta-AI. In parallel, OpenAI is testing advertising in ChatGPT to cover infrastructure costs. While traditional software companies face pressure, a fundamental shift is visible: Investors now evaluate AI systems based on saved salaries, not software costs. The startup InfiniMind is tapping previously unused video and audio data for business analytics.

People

Topics

  • EU Regulation & Digital Markets Act
  • Advertising Models in AI Products
  • Labor Market & Automation
  • Video Analysis & Dark Data

Clarus Lead

The EU Commission is putting pressure on Meta and WhatsApp: The company is said to be excluding third-party AI bots from the WhatsApp Business API, thereby favoring its own Meta-AI. This violates the Digital Markets Act and threatens enforcement measures. In parallel, a pivotal shift in investor logic is becoming visible: AI is no longer measured against saved IT budgets, but against job cuts and wage cost savings. This also explains Anthropic's 350-billion-dollar valuation. While OpenAI tests advertising in ChatGPT and traditional software stocks face pressure, the startup InfiniMind is tapping massive amounts of unused video and audio data for structured business analytics.

Detailed Summary

Regulation Instead of Competition: Meta's WhatsApp Strategy Under DMA Scrutiny

The EU Commission is investigating Meta for systematic access barriers in the WhatsApp Business API. Third-party AI bots are to be disadvantaged while Meta's own AI receives preferential terms. The proceedings represent the first significant application of the Digital Markets Act to AI products. Meta argues that alternative distribution channels exist, but the Commission demands immediate relief of restrictions during the investigation.

Advertising as a Business Model for AI Infrastructure

OpenAI is launching a test phase for context-dependent advertising in ChatGPT. Only free users over 18 are affected; all paid tiers remain ad-free. Ads are oriented toward conversation topics, not personal data. Sensitive areas such as health and politics are excluded. OpenAI cites high infrastructure costs as the reason—a signal that AI operations must become profitable.

Paradigm Shift: Investors Count Jobs, Not Software

A decisive turning point is becoming visible: Venture capital providers are increasingly evaluating AI systems based on saved salaries rather than IT savings. Anthropic's funding round (350 billion dollars) is based on this logic. Agentic AI targets the significantly larger item of wage costs. On the stock market, this manifests through sell-offs in classic software stocks—investors fear that autonomous agents threaten existing business models more fundamentally than traditional software automation.

Unstructured Data Becomes Business Intelligence

InfiniMind (founded by former Google employees) is tapping so-called "dark data"—archived videos, store cameras, TV content—through automatic indexing and queryability. With 5.8 million dollars in seed funding, the company is expanding into the United States. The DeepFrame platform (launching April 2026) promises real-time scene and speaker recognition across large video datasets.

Key Findings

  • Meta Cartel: The EU Commission is for the first time attacking AI access barriers under the Digital Markets Act; Meta systematically favors its own AI in WhatsApp bots.
  • Advertising Instead of Margin: OpenAI monetizes ChatGPT via context-dependent ads—a sign that AI infrastructure is uneconomical without additional revenue.
  • Job Focus: Venture capital valuations are shifting from software to wage cost savings; autonomous agents pose an existential threat to traditional SaaS models.
  • Activating Dark Data: InfiniMind turns unused video and audio data into structured business data through AI indexing.

Critical Questions

  1. Evidence: What specific access denials by Meta does the EU Commission have, and how can these be distinguished from legitimate security policies?

  2. Conflicts of Interest: What economic interests do the investors have who value Anthropic at 350 billion dollars, and to what extent is this valuation based on realistic job cut scenarios or speculation?

  3. Causality: Is the sell-off of classic software stocks causally explained by real business model destruction or by sentiment shifts and rotation into AI stocks?

  4. Feasibility: What technical and legal hurdles would Meta specifically need to lower to comply with the EU's demands without endangering its own AI strategy?

  5. Data Protection & Advertising: How does OpenAI ensure that "context-dependent" advertising is not actually based on profiling via earlier chats if personalization itself can be adjusted by users?

  6. Market Delimitation: Is InfiniMind's dark-data approach truly novel, or are there established competitors (e.g., in the video indexing market) that go unmentioned?

  7. Long-Term Consequences: If autonomous AI agents truly take over entire workflows, what does the tax system look like that compensates for wage cost savings?

  8. Regulation: Should the EU Commission, in parallel with access rules, also establish criteria for "fair AI preference" by platform operators?

Source Index

Primary Source: AI Briefing Daily (10.02.2026) – https://audio.podigee-cdn.net/2349340-m-e13243bfebc770d0a526284ce73ddd4d.mp3

Verification Status: ✓ 11.02.2026

This text was created with the support of an AI model. Editorial Responsibility: clarus.news | Fact-Check: 11.02.2026