What a16z is actually funding (and what it's ignoring) when it comes to AI infra

27 snips

Feb 4, 2026

Jennifer Li, a general partner at Andreessen Horowitz who leads AI infrastructure investments, discusses where a16z is placing its bets. She talks about rebuilding chips, software, and models; the rise of multimodal tools like voice and image; why agents may become practical in 2026; and the urgent gaps in search infrastructure and hiring for AI-native startups.

Ask episode

AI Snips

Chapters

Transcript

Episode notes

INSIGHT

AI Infrastructure Needs A Full Rebuild

Infrastructure for AI must be rebuilt across every layer because existing stacks weren't designed for AI workloads.
Jennifer Li says this retooling spans chips, communication, developer tools, and model layers.

ANECDOTE

Hearing Her Own AI-Cloned Voice

Jennifer Li described cloning her voice with ElevenLabs and hearing herself speak Japanese, which felt uncanny but powerful.
The demo convinced her of rapid progress in audio fidelity and cross-lingual possibilities.

INSIGHT

Agents Hitting Real Productivity

Agents are finally delivering real productivity ROI for longer-running tasks and workflows.
Jennifer Li expects personal and developer agent adoption to surge as the tooling and cloud skills mature.

Get the Snipd Podcast app to discover more snips from this episode

Get the app

Andreessen Horowitz just raised a whopping new $15 billion in funding. And a $1.7 billion chunk of that is going to its infrastructure team, the one responsible for some of its biggest, most prominent AI investments including Black Forrest Labs, Cursor, OpenAI, ElevenLabs, Ideogram, Fal and dozens of others.

A16z general partner with the infra team Jennifer Li (who oversees such investments as ElevenLabs – just valued at $11 billion); Ideagram and Fal, has a clear thesis on where the team is looking to spend it’s latest chunk of cash.

Today on TechCrunch's Equity podcast, Venture and Startups editor Julie Bort talked with Li about where a16z sees this AI super cycle going next, including the talent crunch hitting AI-native startups, why search infrastructure matters more than people think, and what kinds of companies are actually getting funded right now.

Listen to the full episode to hear about:

Where Li thinks the gaps still are when it comes to startups building an AI stack

What makes the most successful AI portfolio companies different

How tools like voice AI are rising in importance (yet still a bit uncomfortable to witness)

The AI startups she's still searching for and is ready to fund

Chapters:

00:00 Intro
01:01 Andreessen Horowitz's $1.7B infrastructure fund
05:00 Crossing the uncanny valley in AI-generated content
07:14 Agents finally becoming real in 2026
09:30 Building your first productivity agent
11:56 Why email agents aren't quite there yet
15:00 Which jobs will agents replace first?
18:05 The most unhinged opinion: Creativity belongs to humans
20:21 The limits of LLMs and the rise of world models
22:13 AI-designed chips are coming
24:00 The truth behind those viral ARR numbers
26:10 Hiring at AI speed: The talent shortage problem
28:47 The pricing mistake that became a big deal
29:21 The future of search for AI agents
30:45 Outro

Subscribe to Equity on YouTube, Apple Podcasts, Overcast, Spotify and all the casts. You also can follow Equity on X and Threads, at @EquityPod.

Learn more about your ad choices. Visit megaphone.fm/adchoices