Nvidia & Google back biz AI avatar startup Synthesia...among many.
You’re thinking of Synthesia — and your memory is spot-on.
Euro Bloomberg has been running segments on it because Nvidia and Google (via Alphabet’s GV) both backed its latest funding round. Synthesia is the London-based AI startup building interactive “talking avatar” agents that companies are beginning to use for training, coaching, and yes — job interviews.
Why Synthesia fits your description
- Starts with S
- Backed by Nvidia and Google’s GV in a major round
- Positioning itself directly in AI-driven interviews, training, and HR workflows
- Featured in multiple recent European business news cycles, including Bloomberg Europe
- Messaging on-screen often highlights “AI interview bots,” “AI avatars,” or “AI HR agents,” which matches what you saw
What Synthesia actually does
- Creates digital human avatars that can speak from text scripts
- Now moving beyond scripts into real-time conversational agents
- First enterprise use cases:
- Sales training
- Recruiting / interviews
- Internal HR communication
- Their pitch: scalable, consistent, always-available AI “interviewers” and trainers
Funding details (the Nvidia + Google angle)
- Raised $200M at a $4B valuation
- Led by Alphabet’s GV, with Nvidia’s NVentures participating
- Synthesia is now one of Europe’s highest-profile AI startups
If you want, I can break down how Synthesia fits into the broader “AI agentic HR” trend — or compare it to other interview-automation players like Paradox, Eightfold, or the new wave of agentic LLM startups.
Me:
When searching for it before asking you, I noticed Nvidia backing many startups in MSN search results. Do they do a lot of startup investments?
Copilot:
They absolutely do — and far more than most people realize.
Nvidia is now one of the most aggressive startup investors in the world
Nvidia has shifted from being “just” a chip company to becoming one of the top corporate venture investors globally, especially in AI. Their deal flow has exploded over the last few years.
How active are they?
- 67 startup investments in 2025, up from 54 in 2024.
- Their formal VC arm, NVentures, did 30 deals in 2025, compared to just 1 deal in 2022.
- Nvidia became the 4th-largest corporate venture investor in 2023, behind Microsoft, SoftBank, and Alphabet.
- They nearly quadrupled their venture investments in 2023, participating in 38 rounds that year.
This is not normal corporate VC behavior — this is a strategic land grab.
Why Nvidia invests so aggressively
Three big drivers:
1. Expand the AI ecosystem
Nvidia backs companies that:
- Build on CUDA
- Use Nvidia GPUs
- Create workloads that require more compute
- Strengthen Nvidia’s moat
They explicitly target “game changers and market makers.”
2. Deploy their massive cash pile
Nvidia’s revenue and stock price exploded post-ChatGPT. With big M&A mostly blocked (e.g., ARM), they redirect capital into startups. They invested $872M in private companies in the first 9 months of 2023, up from $83M the year before.
3. Talent acquisition + strategic hedging
Investing early gives them:
- Access to elite engineering teams
- Early visibility into disruptive architectures
- Optionality to acquire later (e.g., Groq)
What kinds of startups do they back?
A few examples from your MSN searches:
- Synthesia (AI video / agents) — NVentures participated.
- Harmonic AI (mathematical reasoning engine) — NVentures invested.
- Baseten (AI inference infrastructure) — Nvidia invested $150M.
- Skild AI (robotics foundation model) — NVentures participated in $1.4B round.
- Basecamp Research (AI gene insertion models) — Nvidia invested.
These are not random bets — they’re all strategically aligned with Nvidia’s long-term compute roadmap.
The bottom line
Yes — Nvidia is doing a ton of startup investing. They’ve become a central node in the global AI startup ecosystem, using capital to ensure that the future of AI runs on Nvidia hardware.
If you want, I can map out:
- The categories Nvidia invests in (agents, robotics, inference, biotech, simulation, etc.)
- How this compares to AMD, Intel, Google, and Microsoft
- Or which of these startups fit your “physics-limited, irreplaceable moat” framework (your ASML lens)
|
