An interview with Salesforce CEO Marc Benioff on his background and the founding of Salesforce, his vision for AI, Agentforce, data and disruption, and more
a limitless digital workforce. Welcome to Agentforce, now available for all Salesforce customers. Going beyond Copilot, Agentforce combines powerful models with secure, intelligent data [video] Justin Cormack / @justincormack : Then I got so freaked out that I went into Larry Ellison's office and I said, “Listen, think I need to leave the company because there's this huge thing happening in the world and I'm not sure what I'm supposed to do about it, but I know it's something” https://stratechery.com/... Karma / @0xkarmatic : Honestly, this is a good explanation from @Benioff . This is “Founder Mode”. Most of the CEO's are still doing corporate speak and treating LLMs as an oracle which can solve any problem but this was an exception. Marc Benioff / @benioff : Welcome to a new era of ‘agent-first’ companies an era of AI abundance—powered by AI and driven by core values. What if our workforces had not limits? Agentforce is empowering businesses to redefine productivity and customer relationships, paving the way for a new AI-abundant LinkedIn: Brendan Keeler : I haven't always been a fan of the Stratechery interview format, but they've grown on me over time, to the point that they're often the most valuable of the week. …
Context & Ripple Effects
Salesforce had already made a hard pivot toward Agentforce and then rolled out agents for sales, marketing, commerce and customer service. This interview supplies the strategic rationale: Benioff frames the product as an agent-first extension of Salesforce's enterprise software footprint rather than a standalone assistant.
The emphasis on models paired with secure enterprise data explains why Salesforce is positioning Agentforce beyond a copilot. Later coverage of Agentforce 3's observability and MCP support shows that deployment controls and interoperability became central to carrying that positioning into production.
First-order effects
- Salesforce customers can evaluate Agentforce as a broadly available layer for automating work across Salesforce applications, with Salesforce tying its AI pitch to the data already held in those systems.
- Salesforce shifts its near-term product narrative from assistive AI to agents that can act on customer-facing and operational tasks, raising the importance of its data, security and application integration story.
Second-order effects
- CRM buyers must assess agent deployments as workflow and governance decisions, not simply as a feature upgrade; the earlier task-specific Agentforce rollout makes the affected sales, service, marketing and commerce functions explicit.
- Competing enterprise-software vendors face pressure to pair their own AI agents with governed customer data and embedded workflows, rather than compete solely on model access or chat interfaces.
Third-order effects
- If agent use expands inside established enterprise suites, the durable advantage may move toward platforms that combine application context, data controls and operational tooling—not the model alone.
- The pattern points to a more agent-mediated enterprise software stack, where adoption depends on proving reliability, oversight and integration across business processes; the pace remains contingent on customer deployment results.
The trend: Enterprise AI is moving from copilot-style assistance toward embedded agents that execute work within governed business systems.