One-Line Summary

Enterprise bets on AI coding scale up as YouTube adds automatic labels for photorealistic AI videos.

Big Tech

YouTube adds automatic labels for photorealistic AI videos

YouTube will automatically apply AI labels when its systems detect “significant photorealistic AI” and will place those labels more prominently on long-form videos and Shorts so viewers can spot them at a glance. This supplements existing creator self-disclosure and reflects the rise of powerful video models that can generate realistic footage. ¹

Creators can update a video’s disclosure if they believe it was misidentified, but they can’t remove labels when the video was made with YouTube’s own AI tools like Veo or Dream Screen or when C2PA metadata indicates fully AI-generated content. Labels won’t affect recommendations or monetization; for unrealistic or slightly altered content, labels remain in the expanded description, while long-form videos get labels below the player and Shorts get an on-video overlay. ¹

IGN notes the move follows a broader crackdown on misleading AI videos, including takedowns of fake movie trailer channels and pressure from rights holders; the update aims to give viewers clearer context and reduce confusion. For brands and creators, expect more visible disclosures on realistic AI content across channels. ²

Industry & Biz

Cognition raises $1B at $25B pre-money to scale Devin

Cognition, the startup behind the autonomous AI software engineer Devin, raises more than $1 billion at a $25 billion pre-money valuation, led by Lux Capital and General Catalyst with participation from Founders Fund, 8VC, Ribbit Capital, Atreides, and Layer Global. It marks a sharp jump from the company’s $10.2 billion post-money valuation after a $400 million round eight months earlier (Sep 2025). ³

The company says enterprises like Mercedes-Benz, NASA, Goldman Sachs, and Santander are customers, and it claims a $492 million annualized revenue run-rate with enterprise usage growing 50% month-over-month for the past six months. These figures are company-reported and point to rising interest in AI coding agents for real workloads. ³

TechCrunch frames the raise as a vote of confidence that independent AI coding players can coexist with big-platform offerings such as Anthropic’s Claude Code, OpenAI’s Codex, and Google’s efforts following its Windsurf acqui-hire; Cognition itself acquired the remaining Windsurf assets in 2025. For buyers, it means more choice—and more diligence—when comparing independent agents against platform-native tools. ³

Community Pulse

Hacker News (336↑) — Mixed views: some appreciate YouTube’s discovery and context, others worry false positives in moderation could punish honest creators. ⁴

"Their recommendation engine has surfaced awesome obscure content I would have never found otherwise so I value it. Stuff like random recorded conference talks with 3 views. A super enthusiast in Latvia. It does recommend crap sometimes but on balance I like it." — Hacker News ⁴

"Please tell me this is a joke, or that you're not building anything important at work. It's a very well known problem that YouTube's algorithmic moderation hurts a lot of honest creators, and their ability to make a living, when there is a false positive or is abused." — Hacker News ⁴

What This Means for You

For marketers, comms, and creators, AI disclosures on YouTube are moving from “nice-to-have” to “hard to miss.” If your video includes realistic AI elements, viewers will likely see a label directly under the player or on a Short, and labels may appear even if you forget to disclose. Align team workflows so disclosures are added deliberately rather than retroactively. ¹

Brand and legal teams should account for cases where labels become permanent—especially when using YouTube’s own AI tools or when C2PA metadata is embedded—since this affects how audiences perceive authenticity. Plan creative, copy, and influencer briefs with explicit guidance on when and how to disclose realistic AI use. ¹

For product and engineering leaders, Cognition’s funding and customer list suggest enterprises are testing AI coding agents on meaningful tasks. If you explore these tools, define contained use cases, compare against platform-native options, and track concrete metrics like time-to-PR, code review effort, and defect rates to judge ROI. ³

Budget owners should treat vendor selection like any major SaaS decision: request references from named customers, verify security posture, and map how an agent integrates with your repos and ticketing. Independent agents can mean flexibility, but platform tools may reduce integration overhead—evaluate both. ³

Action Items

1. Refresh your YouTube disclosure workflow: For your next uploads, use Creator Studio’s disclosure and verify how labels appear on both long-form and Shorts so your team knows what viewers will see.
2. Decide your metadata policy for AI videos: If you publish AI-generated or altered footage, choose tools that can embed C2PA metadata and document when labels may become permanent.
3. Scope an AI coding agent inquiry: Request information or a demo from Cognition (or a platform alternative), prepare 1–2 low-risk repo tasks, and define success metrics before any broader evaluation.
4. Update brand and influencer briefs: Add clear rules for realistic AI use in sponsored or co-created content, including when to disclose and how to script on-screen clarifications.