Listen to today’s podcast: https://www.youtube.com/channel/UC-nqwUyvLDEvs7bV985k-gQ
AI Daily Podcast 12/15/2025
Today’s podcast episode was created from the following stories:
Figure AI CEO says 170,000+ people applied to his robot company in the last 3 years — fewer than 500 were hired
Figure AI received 176,000 job applications since 2022 but hired roughly 425 people (~0.24%), highlighting both the frenzy for AI/robotics roles and the limits of applicant-tracking systems. CEO Brett Adcock says manual review is still a slog and hinted at building a model to triage resumes. The robotics startup recently raised over $1 billion at a $39 billion valuation.
Harvey CEO shares his advice to young lawyers — and those considering pivoting into legal tech
Harvey cofounder and CEO Winston Weinberg says the fundamentals of great lawyering won’t change with AI; the best partners deeply understand business needs and craft the right story. He urges firms to give juniors more client exposure, even if it means learning from mistakes. Fresh off an $8 billion valuation, Weinberg expects multiple winners in legal tech and advises aspiring founders to build resilience and tolerance for failure.
At this small buyout firm, talking about AI for cost-cutting is off-limits
Tide Rock bans pitches that frame AI as a cost-cutting tool, directing its engineering resources toward growth instead. CEO Ryan Peddycord says in-house capabilities help source deals, identify new customers (like aerospace and defense suppliers), and speed up deployments such as CRM rollouts from months to weeks. The founder-friendly model aims for sustained organic growth and protects the legacy of acquired businesses.
Rivian’s autonomy chief says lidar is very affordable and a no-brainer
Rivian will add lidar to a future version of the R2 SUV as part of its roadmap to full autonomy, with VP of autonomy and AI James Philbin calling the sensor a “no-brainer” as costs fall toward radar-like pricing. The company believes lidar adds robustness and safety compared with camera-only approaches. R2 launches in early 2026 without lidar, with a lidar-equipped model targeted for late 2026.
Gnome is cracking down on AI-generated code after updating its extensions guidelines
The GNOME Project updated its extension review rules to reject submissions that appear largely AI-generated, citing low-quality code, imaginary APIs, and difficulty verifying developer intent. Reviewers want contributors who can justify and explain their code, using AI tools sparingly for learning or completion. The move aims to maintain quality and prevent bad patterns from spreading across the ecosystem.
OpenAI adjusts compensation policy to prevent AI talent outflow
OpenAI is removing its six-month vesting “cliff,” allowing equity to begin vesting immediately rather than after half a year, according to reporting cited by 36Kr. The change underscores intensifying competition for AI talent and the need to reduce friction for new hires. It’s the latest sign that leading AI firms are recalibrating compensation to retain and attract scarce expertise.
Valve’s “Steam Machine” gaming PC will be in trouble if the RAM pricing crisis worsens
Soaring RAM prices—fueled by enterprise AI data center demand and shifts like Micron winding down its Crucial consumer brand—could push the Steam Machine above its hoped-for $650–$750 range. Even with AMD’s custom APU strategy, prolonged memory shortages risk eroding the device’s affordability pitch. If costs persist, Valve may face tough trade-offs to keep its console-like PC compelling for newcomers.
