Jul 14, 2026 · 51m · allin
The Trillion-Dollar Industries AI Is Disrupting: Voice, Law & the End of the Billable Hour
0:00 / 0:00
▶ Watch on YouTube →
gold bands on the timeline = statements, start to end. Hover to read, click to jump. CC turns on captions
In this episode of the All-In Podcast, host Jason Calacanis interviews ElevenLabs CEO Mati Staniszewski and Legora CEO Max Junestrand to explore how vertical AI startups are scaling to hundreds of millions in ARR while disrupting multi-trillion-dollar industries like voice technology and global legal services.
How this conversation actually went
Every chapter scored 0–10 on four independent dynamics. Hover any point for the reasoning behind the score. The hosts hold 41.2% of the talking time here. How this is scored →
The hosts as informed peer 5.1 Guest teaching 4.3 Guest disagreement 1.9 The hosts pushing back 2.7
speaking balance: gold is the hosts, purple is the guest (3 minute bins)
Sharpest disagreement ▶ 48:49 Max rejects general legal model fine-tuning
Max forcefully rejects the premise of building proprietary fine-tuned general legal models, calling it a total waste of time and money.
Hardest push from the hosts ▶ 25:08 Jason challenges Mati on enabling competitorsJason presses Mati directly on whether relying on frontier labs like OpenAI and Anthropic enables the demise of ElevenLabs given their overlapping ambitions.
Biggest teaching moment ▶ 35:37 Max explains law firm pricing cross-subsidizationMax educates Jason on law firm unit economics, explaining that high associate billing rates cross-subsidize partner expertise that is underpriced for critical bet-the-company moments.
The host holds their own ▶ 44:37 Jason details historical legal database digitizingJason showcases deep domain knowledge of legal tech history by describing the double-blind typing and OCR process legacy databases used to digitize case law in India.
the scores for every segment, with the reasoning behind each
| Chapter | Topic | The hosts as informed peer | Guest teaching | Guest disagreement | The hosts pushing back | Why |
|---|---|---|---|---|---|---|
| ElevenLabs' $600M Revenue Ramp and Team Retention Strategy | 4 | 3 | 1 | 1 | Jason asks structured questions regarding ElevenLabs' rapid revenue growth from $100M to $600M ARR. Mati provides detailed metrics on growth velocity and zero attrition among core research talent. | |
| Embedding AI Engineers Across Business Departments | 4 | 3 | 1 | 2 | Jason brings up vibecoding and token maxing, asking how software quality is maintained across teams. Mati explains their organizational structure of embedding AI engineers into non-engineering departments like legal and talent. | |
| Operating Without Product Managers and Deploying Inbound AI Agents | 4 | 4 | 2 | 2 | Jason asks about product managers and management changes, teasing if ElevenLabs fired them all. Mati corrects him by explaining ElevenLabs has never employed product managers, opting for cross-functional builders instead. | |
| Whisperflow, Foot-Pedal Prompting, and AI Interaction Dynamics | 5 | 4 | 1 | 2 | Jason describes his foot-pedal prompting setup and stream-of-consciousness prompting strategy. Mati adds that users communicate with emotional openness to AI agents in sensitive scenarios like debt collection. | |
| Voice IP Protections, Commercial Licensing, and Restoring Speech | 6 | 4 | 1 | 3 | Jason references right of publicity legal limits and shares a personal story of his voice being cloned. Mati outlines ElevenLabs' three-tiered safety framework and highlights paying over $22 million to voice talent. | |
| Navigating Competition with Frontier Model Labs and AGI Progress | 6 | 5 | 3 | 5 | Jason aggressively questions whether partnering with frontier model labs enables ElevenLabs' ultimate demise. Mati defends their model-agnostic platform approach and emphasizes proprietary audio data labeling. | |
| Sponsor Read: Oracle Cloud Infrastructure | 2 | 4 | 2 | 2 | Includes sponsor read and introduction of Legora CEO Max Junestrand. Max clarifies their growth metrics, noting sustained 50% quarter-over-quarter growth. | |
| The End of the Billable Hour and the Transformation of Legal Work | 6 | 5 | 3 | 4 | Jason brings up high hourly billing rates for law firm partners and associates. Max clarifies law firm economics, explaining that high associate billing cross-subsidizes irreplaceable partner counsel during high-stakes litigation. | |
| Cross-Border Legal Data & AI Capabilities | 6 | 4 | 1 | 1 | Jason demonstrates domain knowledge by highlighting state-by-state legal variances such as non-competes in California versus Massachusetts. Max explains how Legora ingests global multi-jurisdictional legal data to assist international operations. | |
| Disruption of Legacy Players (LexisNexis & Westlaw) | 7 | 5 | 2 | 4 | Jason details legacy legal database compilation history, including OCR scanning and double-blind keying in India. Max emphasizes that legal AI search requires 100% comprehensive court data rather than mere power-law top results. | |
| LLM Partnerships & Narrow vs. General Legal Models | 6 | 6 | 4 | 4 | Jason asks if Legora plans to build or fine-tune proprietary legal base models. Max strongly rejects fine-tuning general legal intelligence models as a waste of time and money, arguing instead for highly targeted task models. |
Statements from this episode (30)
ElevenLabs reaches $600 million ARR after rapid revenue scaling
“It took us roughly 20 months to get to the first hundred million in ARR, roughly 10 months to get to 200, five months to get to 300 and that's how we closed, and end of the last year, and now we are at 600.”
ElevenLabs retained 100% of its initial 10 core employees
“From the first 10 people, we had zero attrition. Everybody is, ah, ah, still at the company from those core research and engineering talent building together with us.”
Staniszewski: ElevenLabs embeds software engineers in non-engineering teams like legal and talent
“The way we are structured is A lot of small teams, especially across the product engineering, but also in how we think about go-to-market optimized for specific industries, telco, financial services, healthcare, so every unit is very tightly knit together and …”
Not using AI is a red flag, but overusing it is too
“If you're not using a lot of the coding software, a lot of the co-working software, then you're probably in the wrong spot. If you're using too much of it, that is also a flag because you maybe are not doing that in the right way.”
ElevenLabs CEO Mati Staniszewski says the company has never employed product managers
“We don't have any PMs.”
Staniszewski: Consumers will soon ask directly for AI customer service agents
“Yeah, it's slowly becoming that you will be asking for a give me an AI agent effectively on the call.”
Calacanis: LLMs perform better with long, spoken stream-of-consciousness prompts
“What these LLMs actually do really well with is taking a massive stream of consciousness where you just keep talking and talking and talking, so I'll give it a one to two minute prompt, then I let go, and it has changed everything.”
ElevenLabs powers voice AI for Revolut, Klarna, and PacBank
“We work with a lot of financial services companies Revolut, Klarna, PacBank”
Staniszewski: People disclose sensitive details more openly to AI than humans
“With AI, people are much more open to share what actually happened, give the information, and suddenly this emotional block of, like, in front of other human, I don't want to be able to say all of that, is, is, is very different.”
Staniszewski: ElevenLabs has paid over $22M to voice talent
“Today we paid back over twenty two million dollars back to the community of talent.”
Staniszewski: ElevenLabs, Disney, and Epic Games brought interactive Darth Vader to Fortnite
“Fortnite launched Darth Vader, which people and players could interact with live, in partnership with the estate, in partnership with Disney, so every player, after reaching a certain stage, could have a Darth Vader interact and help you solve the missions.”
Staniszewski: Voice AI performance depends on model architecture over scale
“It's the architecture that matters, not the scale. You really need to change how the model operates.”
Staniszewski: ElevenLabs built an internal team of 1,000+ data labelers
“We build an internal team of over a thousand contractors that label all those audio assets to make them good.”
Staniszewski: ElevenLabs aims for human-level conversational voice fidelity in 2026
“We hope this year we'll do that same thing for voice where any conversation feels like you're speaking with another human.”
Max Junestrand says Leyora achieved 50% QoQ growth for seven quarters
“It's a 50, sustained 50% quarter over quarter for the last seven quarters.”
Leyora scaled from $1M to $150M ARR faster than Sierra
“So I think we actually just became as of the close last week on Tuesday, one of the fastest enterprise company with a direct sales motion to grow from one to 100 and 150. Bidding Sierra with one quarter.”
Max Junestrand states legal market is $1T while software spend is $40B
“It's a trillion dollars every year into legal services, which is very fragmented, but the software spend into legal technology is about forty billion.”
Junestrand: Law firms overcharge for associates to cross-subsidize partners
“In law firms, the way that that business model works is you overcharge for the associates and you actually undercharge for the partners.”
Legora closed a 2026 acquisition in 12 days using its own AI
“We acquired four businesses so far this year. We did the diligence in house the with our own tool and the fastest transaction with it was 12 days. From LOI to closing.”
Junestrand: Kirkland & Ellis partners make $5M-$10M annually in profit
“Per partner, they make it between five and ten million every year in profits.”
Junestrand: Legora gathers legal cases and regulations across all global jurisdictions
“We do the hard work of gathering All the cases, all the legislation, all the regulatory updates for every jurisdiction in the world.”
Junestrand: Legora gives corporate legal teams immediate 80% accurate cross-border responses
“Instead of having to call a lawyer who then knows a lawyer in that region who will respond to the query, they can get an 80% accurate response immediately that they can start working off out of.”
Junestrand: The US legal research market is a duopoly
“The U.S. Is peculiar because it's such a duopoly on legal research.”
Junestrand: AI is moving from augmenting legal work to executing end-to-end tasks
“What's really interesting about, especially the agents following the release of Opus. 4.5 and 4.6 is they can now start to do really intelligent case strategy, and they can actually start to combine the witness statements, the cases, and they can really do end…”
Junestrand: Foundation model providers are not competing in Legora's category
“They are not competing in our product category at all.”
Junestrand: Shallow legal features in general LLMs serve as Legora's pipeline
“So it's actually a big pipeline generator for us.”
Junestrand: Building general intelligence models is a waste of time and money
“I don't believe in Fine tuning or building any general intelligence models. I think that's a total waste of time and money.”
Max Junestrand: Legal AI startups fail on customer distribution, not product development
“There's a lot of legal AI companies and very few are making it through and not because it's hard to build stuff. It's actually quite easy to understand where you can build value, but getting it to the customer is very hard”