The Trillion-Dollar Industries AI Is Disrupting: Voice, Law & the End of the Billable Hour

Eleven Labs (Mati Dubov)

• Revenue Growth & Milestones:

  • Annual Recurring Revenue (ARR) reached $600 million, up from $350 million roughly 20 months prior.
  • Company launched in 2022; took 20 months to reach $100M ARR, 10 months to reach $200M, and 5 months to reach $300M.
  • The company now employs approximately 600 people, maintaining zero attrition among the original 10 founding research and engineering staff.
  • The organization is structured into small, cross-functional teams of 5–10 people focused on specific industries (telco, financial services, healthcare).

• Product & Technology Strategy:

  • Launched the first text-to-speech model capable of sounding human in early 2023, serving as the foundation for a 40+ month product rollout.
  • Eliminates traditional Project Manager roles, embedding engineers within product, legal, and talent teams to automate workflows and ensure AI code security.
  • Achieved a "golden era" for voice interaction where AI agents allow real-time interruption, reducing the awkwardness of traditional IVR systems.
  • Utilizes "Whisper Flow" and hardware pedals to enable stream-of-consciousness prompting, improving context retention and response quality compared to typed prompts.
  • Customers report higher emotional openness when discussing sensitive topics (e.g., missed payments) with AI agents compared to human representatives.

• Market Position & Safety:

  • Addresses voice impersonation risks through a three-part safety framework: tracing all generated content, moderating input at voice and text levels, and releasing public AI-detection tools for open-source models.
  • Established a marketplace for voice creation where voice actors can license their cloned voices, having paid back over $22 million to the community of talent.
  • Secured high-profile partnerships for voice IP, including Matthew McConaughey (8-figure deal for interactive content), James Earl Jones' estate for interactive Fortnite experiences, and Headspace for personalized meditation.
  • Deployed voice restoration technology for individuals who lost their voices to ALS, cancer, or other conditions, including a contract read in the US Congress by Jennifer Wexton.

• Competitive Landscape:

  • Maintains model agnosticism, allowing customers to integrate Anthropic, OpenAI, Google, and open-source models, while competing directly on voice-specific architecture and orchestration layers.
  • Prioritizes specific data quality over scale, employing over 1,000 contractors to label unlabeled audio assets to improve model performance.
  • Aims to build proprietary models for narrow interaction tasks in the future to reduce reliance on frontier model costs, though currently maintains strong partnerships with major providers.
  • Claims to have passed historical AGI tests and believes the new metric for intelligence is whether the system is "10x smarter than every single person on the planet."

Legor (Max Shlonsky)

• Market Opportunity & Traction:

  • Targets a $1 trillion annual legal services market, currently only 4% software-driven compared to 96% manual services.
  • Achieved record growth, becoming one of the fastest enterprise companies to go from $1M to $150M in direct sales within seven quarters (approx. 50% quarter-over-quarter growth).
  • In one quarter, facilitated a direct sales transaction closing in 12 days from Letter of Intent (LOI) to closing, significantly faster than traditional firm timelines.

• Product & Operational Model:

  • Replaces hourly billing with fixed-fee and success-fee models for transactions and litigation, aligning incentives to close deals rather than extend them.
  • Deploys "Legal Engineers" to assist law firms in transitioning from document-heavy workflows to AI-agentic orchestration.
  • Aggregates structured data on global legislation, cases, and regulatory updates to provide localized legal analysis across different jurisdictions (e.g., California non-competes vs. South African law).
  • Utilizes narrow, fine-tuned models for specific tasks like "tabular review" to reduce latency and costs, avoiding the expense of building general-purpose legal intelligence models.

• Competitive Dynamics & Strategy:

  • Competes with legacy players like LexisNexis and Westlaw, which struggle with pivoting to AI-native operations and face valuation pressures due to AI disruption.
  • Views competitor AI offerings (e.g., Anthropic's legal tools) as a pipeline generator that validates the market before customers hit scalability ceilings and switch to Legor.
  • Maintains strict data security and compliance as a core currency, hosting data for national secrets and weapons manufacturers without requiring on-premises deployment.
  • Pursues an M&A strategy to integrate existing content providers and expand into smaller jurisdictions where legacy players lack agility.

• Future Outlook:

  • Anticipates a shift where AI moves from augmenting legal work to performing end-to-end tasks (combining witness statements, cases, and strategy), changing the junior lawyer role to focus on agent orchestration.
  • Believes the future of legal research requires 100% comprehensive data coverage rather than top-tier subsets, necessitating the manual digitization efforts currently being scaled by AI.