For much of the AI boom, software investors have worried about one uncomfortable possibility:
What if artificial intelligence makes traditional software less valuable?
Salesforce's latest earnings offer a very different scenario.
Shares jumped roughly 14% in after-hours trading after the company reported stronger second-quarter results, raised its fiscal 2027 outlook and unveiled an expanded partnership with Anthropic.
The immediate catalyst was earnings.
The more important story may be that Salesforce is beginning to show how an established software platform can monetize AI rather than simply defend itself from it.
Salesforce's AI Business Is Becoming Material
Salesforce reported fiscal Q2 revenue of $11.35 billion, up 11% year over year.
Current remaining performance obligations, an important measure of contracted future business, increased 14% in constant currency to $33.5 billion.
But the most striking figure was Salesforce's AI and data business.
Agentforce and Data 360 ARR reached nearly $3.9 billion, up more than 210% year over year.
Agentforce ARR alone exceeded $1.5 billion, growing more than 240%.
Those numbers matter because the AI debate surrounding software has often been highly theoretical.
Salesforce is beginning to show measurable recurring revenue attached to AI products.
That changes the conversation.
Claudeforce May Be More Important Than the Earnings Beat
On the same day, Salesforce and Anthropic expanded their partnership with a product called Claudeforce.
The integration connects Claude with Salesforce's data, workflows, business logic, actions and governance systems.
The initial release includes 37 prebuilt sales skills designed to let users and AI agents reason over live CRM information, update pipelines and take controlled actions directly from Claude.
This illustrates a potential competitive advantage for established enterprise software platforms.
Large language models can reason.
But reasoning alone does not give an AI system access to a company's customer records, permissions, transaction history, workflows or business rules.
Those assets often live inside systems such as Salesforce.
The implication is important:
> AI models may become the interface, while platforms such as Salesforce remain the system of record and execution layer underneath them.
If that architecture becomes common, AI does not necessarily eliminate Salesforce.
It may increase the value of the data and workflows Salesforce already controls.
The Market Is Starting to Separate Software Winners and Losers
This is particularly interesting because software stocks have reacted very differently to AI.
Intuit recently came under pressure after weaker guidance revived concerns that generative AI could disrupt parts of traditional financial and productivity software.
Salesforce is presenting the opposite case.
Its argument is effectively that enterprises need AI to work through trusted corporate systems rather than around them.
Salesforce reported 7 billion Agentic Work Units delivered across Agentforce and Slack, including 3.2 billion during Q2 alone, up 97% sequentially.
That metric is company-defined, so investors should not treat it like standardized revenue.
But it does show a rapidly increasing volume of agent activity moving through Salesforce's ecosystem.
Guidance Improved Too
Salesforce raised its full-year fiscal 2027 revenue outlook to $46.1 billion to $46.4 billion, representing growth of approximately 11% to 12%.
The increase includes contributions from pending Contentful and Fin acquisitions as well as roughly $100 million of incremental organic growth, partially offset by foreign-exchange pressure.
That nuance matters.
Not all of the guidance increase comes from organic acceleration.
But Salesforce is simultaneously showing stronger AI ARR, improving bookings and a faster pace of contracted demand.
The combination is much more compelling than an AI narrative unsupported by financial results.
The Risk: AI May Strengthen Competition Too
The bullish interpretation is not guaranteed.
If AI makes it easier for competitors to build CRM functionality, parts of Salesforce's traditional product suite could face pricing pressure.
Enterprises may also become less willing to pay for numerous individual software seats if autonomous agents can perform work previously done by employees.
That could ultimately reshape the economics of per-seat SaaS pricing.
Salesforce therefore needs to prove that Agentforce revenue can grow faster than AI compresses older software economics.
The company is clearly trying to get ahead of that transition by shifting the value proposition away from individual software interfaces and toward data, governance, workflow and agent execution.
Why This Earnings Report Matters
Salesforce's quarter is important because it challenges the simplistic idea that AI automatically destroys software companies.
The more useful distinction may be:
Which software companies own data and workflows that AI needs to access?
Companies whose primary value is a basic interface may face disruption.
Companies controlling mission-critical enterprise systems may be able to charge for the infrastructure AI agents need to operate safely.
Salesforce is trying to establish itself firmly in the second group.
That is why the stock's post-earnings surge is about more than one quarterly beat.
The market is beginning to price a possibility many software investors had started to doubt:
Some of the largest SaaS companies could become major AI beneficiaries rather than casualties.