Metaview's €53M Bet on Agentic Recruiting

Metaview's €53M Bet on Agentic Recruiting

London's Metaview raised €53.1M ($60M) Series C led by Insight Partners on September 30, pushing total funding to €97.4M[5]. The platform doesn't just assist recruiters — its "fillmore" agent autonomously sources, screens, and interviews candidates. It already serves 7,000+ companies, including Deel and Affirm, and is using the raise to grow headcount to 250 and open a New York office[6].

This matters beyond HR tech. Metaview is a clean example of a company that skipped the "AI copilot" phase entirely and went straight to autonomous agents doing full workflows end-to-end. Investors and HR tech voices on X framed this as validation that agentic products — not just chat assistants bolted onto existing software — are where the capital is flowing now.

The lesson for builders: if your product still has a human in the loop for every step, you're building last year's company.

The EU AI Act Is Already Live for Agents — Whether You Noticed or Not

A wave of September analyses confirmed what should have been obvious: AI agents are already covered by the EU AI Act under existing AI system and GPAI definitions — there's no separate "agent" carve-out coming[7]. Article 50 transparency obligations have been active since August 2026. The scarier deadlines (high-risk obligations) land December 2027 for standalone systems and August 2028 for embedded ones, with a new EN 18286:2026 standard now offering Article 17 compliance guidance[8].

The sharper point buried in this: prompt injection vulnerabilities are being treated as a direct path to high-risk non-compliance. That's a security problem and a legal problem fused into one. Policy and security researchers on X spent the week debating whether self-regulation can actually hold here, or whether agentic systems need statutory oversight baked in from day one — Nordic and EU voices leaned hard toward the latter.

If you're shipping agents into the EU market and treating this as a 2027 problem, you're already behind. Transparency obligations are live now.

Voice Gets a Real Upgrade: OpenAI's GPT-Live-1

OpenAI quietly shipped something that'll matter more than it got credit for: GPT-Live-1, a real-time bidirectional voice API with custom voices and telephony support[9]. This isn't another TTS wrapper — it's full-duplex, ambient conversation infrastructure, built for the interruption, backchanneling, and pacing of how humans actually talk, not scripted turn-taking.

Developers on X were visibly excited, specifically about use cases like spam-call filtering, natural phone ordering, and agents that can hold a real conversation on the phone without the uncanny pause-and-respond rhythm that's plagued voice AI since the start.

For anyone building voice products — including us — this closes a gap that's been the single biggest tell that you're talking to a machine. Full-duplex is the unlock.

Metaview's Raise Is a Signal for the Whole Region

Metaview's round, while headquartered in London, is being read across European tech circles as confirmation that the continent's AI agent and infrastructure plays can still pull serious international capital — Insight Partners led, but the deal flow and founder base skew heavily European[5]. Nordic and EU builders on X framed this as proof that "trusted, rules-based AI" isn't a competitive disadvantage against Silicon Valley speed — it's becoming a selling point, especially for enterprise buyers who now have to think about AI Act compliance regardless of where their vendor is based.

That's the real throughline connecting Metaview to the EU AI Act story: Europe isn't trying to out-release the US labs. It's betting that governance maturity becomes a product feature once enterprises actually have to answer to regulators. The companies building compliance into the architecture now, rather than bolting it on in 2027, are the ones who'll win enterprise trust first.