AI Is Reshaping SaaS: How Claude Co‑Work and Agentic Tools Are Creating a New Software Paradigm

In this article, we draw directly on insights shared by David Sacks, Brad Gerstner, and David Friedberg on Episode 260 of the All-In Podcast (Spotify link). Their discussion explores how agentic AI is shifting valuations, business models, and competitive strategy in the SaaS world.

The release of Claude Co‑Work, an agentic AI capable of orchestrating workflows across tools and data, sparked a dramatic market reaction earlier this year. Software stocks sold off sharply as investors began to price in risks that agentic AI could erode long-standing SaaS moats and upend traditional business models.

For decades, SaaS companies were valued in part because each new user seat represented predictable, recurring revenue. AI now threatens that assumption, prompting a fundamental rethinking of how software businesses generate and capture value.

The Valuation Shock: Why Markets Are Discounting SaaS

The release of Claude Co‑Work sent a jolt through SaaS markets. Within days, companies like Thomson Reuters, LexisNexis, LegalZoom, Figma, and Salesforce experienced double-digit drops in stock value, with some losing billions in market capitalization almost immediately.

Investors are reacting not because revenue is falling but because AI introduces uncertainty about future cash flows. Brad Gerstner described this as a “forward discount,” explaining that investors are questioning whether the high multiples they are paying today make sense when AI can automate work previously performed by humans, potentially reducing the number of seats and diminishing future recurring revenue.

Two Forces Eroding Traditional SaaS Revenue

1. Internal AI Replacements For simpler SaaS tools, organizations now have the option to create internal versions powered by LLMs connected to their own databases. While maintenance needs make large-scale adoption unlikely, the mere possibility introduces risk, causing investors to discount future cash flows.

2. Reduced Human Dependency and Seat Count Collapse Agentic AI can now perform multiple downstream tasks directly, meaning customers may require fewer employees to achieve the same outcomes. Fewer users translate into fewer software seats, directly affecting revenue projections. 

Both forces make future cash flows less predictable, which is the core reason for the multiple compression observed in SaaS valuations.

The Structural Threat: Value Migrating to the Agent Layer

The deeper competitive threat is that value will migrate beyond the SaaS layer to agentic AI that orchestrates work across systems. Agents like Claude Co‑Work can:

• Pull data from multiple tools and databases

• Automate cross-platform workflows

• Serve as the primary workspace for task completion

Users increasingly prefer a single AI workspace that spans all tools, reducing the need to navigate multiple SaaS UIs. When value migrates to this agent layer, underlying SaaS platforms risk becoming systems of record (the data silo) rather than systems of action (where work actually happens).

The Data Strategy: Open vs. Closed Ecosystems

To prevent themselves from being rendered into systems of record, many SaaS organizations are deliberately limiting data accessibility. This buys them time to build their own proprietary AI features.

However, this "lock-in" strategy is risky. Modern businesses want tools that work together, and they are increasingly moving away from closed systems toward open ones. Paradoxically, the SaaS tools that make their data easiest for AI agents to use may become the most "sticky" by embedding themselves as the essential foundation for a company’s entire AI workflow.

The Counter-Argument: Why SaaS Isn’t Dead

Despite the alarmism, many SaaS platforms remain remarkably resilient. Established platforms possess structural protections that AI cannot easily bypass:

• Entrenched Complexity and Data Lock-in: Large, complex SaaS platforms, such as Salesforce and SAP, that manage millions of customer contacts and have evolved their offerings over decades of bug fixes and feature iterations, remain relatively protected. These enterprise systems, including CRMs and ERPs, have layers of integrations, data models, and workflows that are difficult to replace or replicate, even with advanced AI.

• The Governance & Liability Gap: Established SaaS vendors provide the compliance, permissioning, and data provenance that large organizations require for legal and regulatory reasons. AI agents currently lack the "accountability layer" that enterprise procurement demands.

• The High Cost of Internal Maintenance: While companies can build internal AI versions of SaaS tools, the long-term cost of maintenance, security patching, and API management often exceeds the cost of a subscription.

The SaaS Evolution: From Selling Tools to Selling Outcomes

Here is the catch: Being "un-replaceable" doesn't protect your revenue. Even if a company stays with a complex platform like Salesforce, it will still buy fewer seats if its employees become 10x faster because of AI. This "efficiency gap" is forcing a fundamental shift in how software is sold.

Instead of paying for "access" to a tool (the seat), customers will increasingly pay for completed work (the outcome).

• Old Way: Pay $100/month for a seat so a human can write marketing emails.

• New Way: Pay $10 for every AI-generated, high-conversion email campaign the software completes for you.

In this new paradigm, the value isn't in the tool; it's in the result. Companies that adapt early to this shift will capture value in fundamentally new ways. SaaS organizations need to act on multiple fronts:

• Rethink monetization models: transition from seat-based to usage- or outcome-based models.

• Invest in open data strategies: Don't just build for humans; build your software so that AI agents can be your power users.

• Focus on value orchestration: provide AI-enabled insights and cross-workflow capabilities.

• Reframe AI internally: use agentic AI to enhance platform value rather than see it as a competitor. 

Is Your SaaS Strategy Ready for the Agentic Era?

At OpsGuru, we help technology companies and ISVs modernize their stacks to capture value in this new paradigm and thrive with new technologies like agentic AI. From data readiness to AI-integrated cloud infrastructure, let’s build what’s next.

Explore OpsGuru’s SaaS & ISV Solutions