Databricks CEO: AGI Is Already Here, Why Companies Are Behind, and the Superintelligence Debate
Databricks CEO: AGI Is Already Here, Why Companies Are Behind, and the Superintelligence Debate
Podcast1 hr 2 min
Listen to Episode
Note: AI-generated summary based on third-party content. Not financial advice. Read more.
Quick Insights

The interview offers no high-conviction stock trade or price target; Databricks, Anthropic, and OpenAI are private, while the public companies mentioned lack specific buy or sell analysis. For public-market exposure, monitor Snowflake (SNOW) as a data-platform peer, but treat it as a watchlist idea—not a buy signal—without valuation or comparative analysis. Consider AI enterprise data infrastructure and cybersecurity as longer-term themes, while accounting for gradual adoption, rising AI costs, and security risks. Treat AI drug discovery partnerships involving Regeneron (REGN) and Merck (MRK) as developments to watch, not proven investment catalysts; clinical trials remain a key hurdle.

Detailed Analysis

Databricks (Private)

  • The host described Databricks as a private company with a reported $7 billion revenue run rate and 80% year-over-year growth; those figures were presented as a rumor, not confirmed by the CEO.
  • CEO Ali Ghatzi said the company helps organizations manage and use data for AI, including through governance and security tools. He described AI agents as a potential source of increased Databricks usage and said the company charges based on consumption.
  • Ghatzi said Databricks is expanding into security and marketing. He also said he wants growth to exceed the current 70%–80% range.
  • On going public, Ghatzi cited better access to capital and visibility as advantages. He said remaining private avoids public-market volatility and distraction, and noted that the company provides employees with periodic liquidity through tenders.

Takeaways

  • Databricks offers exposure to enterprise data infrastructure and AI adoption, but it is not publicly traded, so the discussion does not provide a direct public-stock opportunity.
  • The business case depends in part on enterprises moving beyond basic chatbot and coding uses toward broader agent adoption. Ghatzi said that rollout could take 10–15 years and that organizational barriers remain substantial.

Snowflake (SNOW)

  • Snowflake was named as a competitor to Databricks. The discussion did not compare the companies’ products or financial results in detail.

Takeaways

  • Snowflake is a publicly traded way to follow the data-platform theme raised in the interview, but the transcript offers no specific buy or sell recommendation and no comparative valuation analysis.

Anthropic (Private)

  • Anthropic was cited as an AI provider whose products are expanding into areas such as spreadsheets and image editing.
  • Ghatzi said AI providers are doing important work, but argued that enterprises may worry about vendor lock-in and retaining control of their data and intellectual property when using external AI platforms.

Takeaways

  • The discussion highlights both the opportunity in rapidly expanding AI products and the potential importance of data control and integration for enterprise customers. Anthropic was not presented as a public-stock opportunity, and no investment recommendation was made.

OpenAI (Private)

  • OpenAI was mentioned alongside Anthropic as a provider of AI platforms that organizations may use.
  • Ghatzi’s broader point was that companies need to address security, cost, access to organizational context, and control of intellectual property before they can deploy AI widely.

Takeaways

  • The interview suggests that AI model providers are only one part of the enterprise adoption opportunity; tools that help organizations deploy models securely and cost-effectively may also benefit. No specific investment recommendation was given.

Microsoft (MSFT)

  • The hosts noted that Anthropic was expanding into products that could compete with Microsoft, including an Excel-like product.
  • The conversation framed this as an example of AI companies moving quickly into established software categories.

Takeaways

  • The discussion points to competitive pressure on established software companies as AI products expand into existing markets. It does not assess Microsoft’s financial exposure or provide a price target or recommendation.

Cloudflare (NET)

  • Cloudflare was mentioned as another company evaluating products in adjacent markets as AI makes software development faster.

Takeaways

  • The interview highlights how AI may intensify competition across software categories. It does not describe a specific Cloudflare product or make a company-specific investment case.

Block (XYZ; formerly Square)

  • Ghatzi said digital-native companies may be further along in adopting AI. He cited Block and Jack Dorsey’s “company world model” as an example of thinking about how AI can understand a company’s operations.

Takeaways

  • Block was presented as an example of potential AI adoption inside a company, not as a stock recommendation. The discussion did not provide financial details or a timeline for benefits.

Airwallex (Private)

  • The episode’s sponsor described Airwallex as an AI-native platform for global accounts, cards, and payments, contrasting it with legacy financial infrastructure.

Takeaways

  • The ad points to a potential theme in modernizing cross-border payments and financial services. These claims came from promotional material, and the episode did not provide independent analysis or an investment recommendation.

Regeneron (REGN)

  • Ghatzi said Databricks worked with Regeneron researchers on a project that identified a genome associated with chronic liver disease and helped support drug discovery.
  • He described AI as a way to accelerate research, while noting that treatments still need to go through human trials.

Takeaways

  • The example illustrates a possible use of data and AI in biotechnology research. The transcript does not identify a specific drug, clinical milestone, or investment recommendation for Regeneron.

Merck (MRK)

  • Ghatzi said Databricks trained a model with Merck called TEDDI—Transformer Enabled Drug Discovery. He described it as analyzing gene-regulatory networks to help distinguish disease causes from reactions to disease.
  • He presented the model as a potential way to accelerate drug discovery, but did not name a specific drug or clinical outcome.

Takeaways

  • AI-enabled drug discovery is a theme to watch, but the discussion does not establish that TEDDI has produced a marketable treatment. Drug-development progress and human trials remain important uncertainties mentioned in the conversation.

Netflix (NFLX)

  • Ghatzi said Databricks’ early research at UC Berkeley included a recommendation algorithm intended to help Netflix recommend movies. The interview also discussed Netflix’s historical prize for improving its recommendation algorithm.

Takeaways

  • The example shows how data-driven recommendations can support consumer internet businesses, but it is historical context rather than a current Netflix investment thesis.

Alphabet / Google (GOOGL, GOOG)

  • Google was mentioned as an example of a company using machine learning, and its search experience was used as a comparison for current enterprise AI use: many companies still use AI mainly as a more capable search tool.

Takeaways

  • The discussion raises the possibility that AI could change how users find information, but it does not assess the impact on Google’s business or make a stock recommendation.

Meta Platforms (META)

  • Facebook was cited as an early example of machine learning being used in advertising.

Takeaways

  • This was historical context about machine learning adoption, not a current investment analysis or recommendation for Meta.

Uber (UBER)

  • Uber was cited as an example of a company using machine learning for routing.

Takeaways

  • The example illustrates the value of applying AI to operational data, but the transcript provides no current Uber-specific outlook or investment recommendation.

AI Enterprise Adoption and Data Infrastructure (Theme)

  • Ghatzi said many organizations still use AI mainly for search and coding, with coding gains often around 20%–30% in the customer feedback he described.
  • He identified four enterprise challenges: getting relevant organizational context into AI systems, protecting data and systems, controlling token costs, and avoiding excessive dependence on outside providers.
  • He argued that AI agents could reduce delays caused by organizational handoffs, but said broad adoption is constrained by legacy systems, compliance, security, and internal processes.
  • He said Databricks sees agents generating more queries and consumption, but also acknowledged that AI costs can rise faster than company revenue.

Takeaways

  • The investment opportunity discussed is broader than AI model development: data platforms, governance, security, and cost management may be important as organizations deploy AI.
  • Adoption may be gradual. The transcript specifically identifies rising AI costs, cybersecurity concerns, and organizational resistance as constraints; it does not offer a near-term revenue or market-size forecast.

Cybersecurity (Theme)

  • Ghatzi described cybersecurity as a real risk as AI agents gain access to more company data and systems. He emphasized the asymmetry between attackers, who may succeed with one intrusion, and defenders, who need to prevent intrusions consistently.
  • He said Databricks offers Lakewatch, a security product that uses data and agents to monitor for intrusions and risks involving agents.
  • He also argued that AI tools can help scan new code for vulnerabilities and that organizations could address weaknesses in older code over time.

Takeaways

  • The discussion supports cybersecurity as a relevant investment theme alongside AI adoption, particularly as companies connect more data and tools to agents.
  • The transcript highlights cybersecurity exposure but does not compare security vendors or provide a specific stock recommendation.

AI Drug Discovery and Healthcare (Theme)

  • Ghatzi described AI as potentially useful in identifying disease-related genetic mechanisms and accelerating drug discovery.
  • He cautioned that claims about curing diseases such as cancer within a specific short timeline are difficult to assess because treatments must still pass through human trials.
  • He also cited applications in detecting self-harm risk among young people and supporting delivery operations, but did not discuss these as investment opportunities in specific public companies.

Takeaways

  • AI-enabled healthcare research is a potential long-term opportunity, but the interview provides examples rather than evidence of guaranteed clinical or commercial success.
  • Human trials were explicitly identified as a constraint; the conversation offered no specific drug-development price targets or timelines.
Ask about this postAnswers are grounded in this post's content.
Episode Description
(0:00) Ali Ghodsi joins the show: Databricks' early bets on cloud, AI, and open source (8:50) Is AGI Already Here? Why Enterprise AI Is Falling Behind (22:04) Live Demo: How Databricks Uses Agentic AI For Efficiency (38:25) Databricks IPO: The Case for Staying Private (44:02) RSI, Cyber Risk & the Doom Debate   Thanks to our partner for making this possible:   Airwallex is a leading global payments and financial platform for modern businesses, offering trusted solutions to manage everything from business accounts, payments, treasury, and spend management to embedded finance. Learn more at https://airwallex.com/allin   Follow Ali: https://x.com/@alighodsi   Follow the besties:  https://x.com/chamath https://x.com/Jason https://x.com/DavidSacks https://x.com/friedberg   Follow on X: https://x.com/theallinpod   Follow on Instagram: https://www.instagram.com/theallinpod   Follow on TikTok: https://www.tiktok.com/@allin   Follow on LinkedIn:  https://www.linkedin.com/company/allinpod   Intro Music Credit: https://rb.gy/tppkzl https://x.com/yung_spielburg   Intro Video Credit: https://x.com/TheZachEffect   #allin #tech #news
About All-In with Chamath, Jason, Sacks & Friedberg
All-In with Chamath, Jason, Sacks & Friedberg

All-In with Chamath, Jason, Sacks & Friedberg

By All-In Podcast, LLC

Industry veterans, degenerate gamblers & besties Chamath Palihapitiya, Jason Calacanis, David Sacks & David Friedberg cover all things economic, tech, political, social & poker.