Deskless workers are the next frontier of tech

3 trends in startups/tech/venture

👷🏽 Deskless workers are the next frontier of tech

• 80% of the global workforce are “deskless” (i.e., don’t work at a desk or in an office). Deskless workers include healthcare providers, retail employees, manufacturing workers, and field service technicians.

• Startups historically overlooked these workers, opting instead for white collar employees working on spreadsheets, slide decks, and coding terminals.

• However, startups focused on serving this overlooked segment are increasingly seeing exits and up-rounds, which includes some of our portfolio:

  • Convex (acquired by ServiceTitan), which identifies upsell opportunities for commercial contractors,

  • Marble (acquired by Caterpillar), which leverages robotics to augment construction and mining operations, and

  • UpKeep (Series B), which leverages real-time mobile data to help maintenance and reliability teams run factories more efficiently.

🤖 AI is increasingly used to augment, not replace

• Microsoft's AI recently successfully helped diagnose cases that stumped experts and doctors with 85% accuracy by teaching AI to think like a panel of doctors rather than as a single expert.

• The news isn’t all good, though. Stanford researchers found that ChatGPT and existing commercial therapy bots not only expressed stigma toward those with mental health conditions but even encouraged delusional thinking.

• These stories show the importance of leveraging AI to augment rather than replace humans. In my role as a voting Commissioner on the California Mental Health Commission, we launched a $100M Innovation Partnership Fund to invest more in mental health solutions — focused on augmentation.

💰 1/3 of all US venture money went to just 16 companies last quarter

• Half of all venture funding went to AI companies in Q2. The largest round? Meta’s investment into Scale AI, representing more than 1/3 of all AI funding.

• I recently shared the stage with Vijay, Field CTO of Scale AI, where we discussed how investments are concentrating in companies with data moats. Startups that collect creative / weird / hard-to-access data can better differentiate their products, and can leverage their unique data to create virtuous cycles for product development. (You can find other takeaways from the conversation here and the full recording here.)

2 theses on what’s next

🤝 Startups win through trust, not just speed

• With AI enabling startups to build and sell faster than ever before, it’s no longer possible to win just with speed. Speed without trust undermines long-term credibility. Trust drives long-term retention and loyalty.

• As the saying goes, “people don't care how much you know until they know how much you care.” The founders who can articulate why they care will be the ones that earn the trust of customers.

♡ The best startups are different, not just better

• AI has lowered the barriers to entry and crowded every product category with competitors vying for market share. Offering a faster, cheaper, or better solution is no longer enough for founders who want to stand out. Founders also need to be different from what already exists.

• Different can include being opinionated (e.g. taking a stance on the best way to solve a problem), tasteful (e.g. designing a differentiated customer experience), or novel (e.g. packaging unexpected solutions together).

1 tool I love

🎓 AI assistant for therapists

• Klarify helps therapists minimize admin work and help therapy clients maximize therapeutic outcomes.

• In addition to automating note taking in therapy sessions, Klarify also gives clients personalized tools for self-reflection and growth between sessions.