Not all software will be 'eaten' by AI: Daniel Newman

Key Concepts

• AI Disruption in Software: The central theme revolves around the market’s overreaction to the potential of Artificial Intelligence (AI) to disrupt the software industry.
• Enterprise Software Resilience: Core enterprise applications (SAP, Oracle, Salesforce, ServiceNow) are considered more resilient to AI disruption due to their complexity, proprietary data, and critical role in business operations.
• Vulnerability of Niche Software: Software focused on specific tasks (project management, expense reporting, design) are seen as more vulnerable to AI-powered alternatives.
• Consumption-Based Pricing: A shift towards pricing models based on usage rather than traditional licensing is highlighted as a positive indicator for companies like Salesforce.
• AI Security Limitations: Current AI security solutions are demonstrated to be fallible, with examples of successful attacks despite AI defenses.
• “Throwing the Baby Out with the Bathwater”: The market is accused of excessively punishing software stocks without differentiating between companies likely to succeed and those at risk.

Software Stock Market Reaction to AI: A Detailed Analysis

The discussion centers on the significant downturn in software stock prices, with 36 stocks experiencing substantial declines despite potentially strong underlying fundamentals. The core argument is that the market is overreacting to the perceived threat of AI, “throwing the baby out with the bathwater.” The speakers emphasize a nuanced view, asserting that not all software will be replaced by AI, and that a distinction must be made between different types of software companies.

The Two-Tiered Software Landscape

A key point is the differentiation between two tiers of software companies. The first, and more vulnerable, tier consists of companies specializing in niche applications like project management tools, design platforms, and expense reporting software. These are considered at risk because AI could potentially replicate or replace their functionalities. The speaker specifically cites these types of companies as being “worried about.”

The second tier comprises large enterprise software providers like SAP, Oracle, ServiceNow, and Salesforce. These companies are deemed more resilient due to several factors:

• Complexity: They provide core applications that manage complex business processes.
• Proprietary Data: They handle sensitive, proprietary data crucial to business operations.
• Global Reach & Compliance: They facilitate transactions across borders, navigating diverse regulatory landscapes.
• Critical Infrastructure: They are integral to the functioning of entire organizations.

AI Security Concerns & Real-World Examples

The discussion highlights the limitations of current AI security solutions. An example is provided of Anthropic, an AI company, being subjected to a massive attack (24,000 Chinese bots) that its AI-powered security software failed to prevent. This illustrates that the idea of AI completely replacing cybersecurity is premature. CrowdStrike, a cybersecurity firm, is noted as being down 11% despite the CEO’s assertion to Claude (an AI assistant) that AI wouldn’t eliminate the need for cybersecurity.

Further emphasizing this point, a story is recounted about a medical lab safety employee granting access to “OpenAI” (likely a misunderstanding referring to access via an OpenAI-powered tool) and subsequently deleting all her emails, highlighting the risks associated with integrating AI tools without proper safeguards. The speaker emphasizes that “these tools are not ready for prime time.” The concept of “agents” – nameless, faceless, agile employees with broad system access – is introduced as a significant security concern.

Salesforce & ServiceNow: The Importance of Delivery & Consumption-Based Models

The conversation turns to Salesforce and ServiceNow, noting their past experiences with AI-related stock dips and subsequent rebounds. The key takeaway is that these companies need to consistently deliver results quarter after quarter to regain investor confidence. Specifically, Salesforce’s “Rule 55” team’s pivot to a “consumption-based agent platform” – one that is more data-centric and leverages proprietary data – is seen as a positive development. The shift towards consumption-based pricing (paying for usage) is also highlighted as a favorable trend.

Market Sentiment & Investment Opportunity

The speakers agree that the market is currently in a “negative growth moment” regarding software stocks, creating a potential opportunity for patient investors. They believe that companies that can successfully integrate AI with their existing critical applications will ultimately thrive. The expectation is that smaller, less resilient companies will likely fade away, while the larger players will adapt and maintain their dominance.

Specific Stock Mentions & Recommendations

Oracle is specifically mentioned as a potentially attractive investment. Conversely, Workday and other stocks recently downgraded by Jefferies are cautioned against. The speaker notes the irony of these downgrades occurring simultaneously with the broader market panic.

Notable Quotes

• “We’re throwing the baby out with the bathwater.” – Emphasizing the market’s overreaction.
• “Not all software will be eaten by AI, and not all software will survive the AI flickr.” – Highlighting the nuanced impact of AI.
• “Businesses operate, they report to the street, they take critical data and transact. They’re going to be collaborators.” – Underscoring the continued importance of established software systems.
• “It’s AI plus critical applications.” – Summarizing the successful formula for software companies in the age of AI.

Technical Terms & Concepts

• SaaS (Software as a Service): A software distribution model where applications are hosted by a vendor and made available to customers over the internet.
• Proprietary Data: Data owned exclusively by a company and considered a valuable asset.
• Regulatory Compliance: Adherence to laws and regulations governing business operations.
• AI Agents: Autonomous entities powered by AI, capable of performing tasks and making decisions.
• Consumption-Based Pricing: A pricing model where customers pay based on their actual usage of a service.
• AGI (Artificial General Intelligence): A hypothetical level of AI that possesses human-level cognitive abilities.

Logical Connections

The discussion flows logically from the initial observation of the software stock downturn to a detailed analysis of the factors driving this reaction. The differentiation between enterprise and niche software provides a framework for understanding which companies are most vulnerable. The examples of AI security failures and the discussion of Salesforce’s strategy reinforce the argument that AI is not a simple replacement for existing software but rather a tool that must be integrated thoughtfully.

Data & Statistics

• 36 Software Stocks: The number of software stocks experiencing significant declines.
• 11% Decline: The current drop in CrowdStrike’s stock price.
• 24,000 Chinese Bots: The number of bots involved in the attack on Anthropic’s AI security software.

Conclusion

The primary takeaway is that the market’s reaction to AI’s potential impact on the software industry is excessive. While AI will undoubtedly disrupt the landscape, core enterprise software providers with complex systems, proprietary data, and established customer bases are likely to remain resilient. Investors should exercise patience and focus on companies that can successfully integrate AI with their existing offerings, rather than assuming a complete overhaul of the software ecosystem. The current downturn presents a potential opportunity for those willing to look beyond the headlines and identify companies poised for long-term success.