Salesforce plans $300 million Anthropic spend as Benioff bets big on AI coding agents

3 Sources

Share

Salesforce CEO Marc Benioff revealed the company will spend $300 million on Anthropic tokens in 2026, almost entirely for coding applications. The investment signals a major shift in enterprise AI spending as coding agents deliver efficiency gains that make software development faster and cheaper. Benioff also teased upcoming AI-driven coding functionalities inside Slack.

Salesforce CEO Marc Benioff Commits $300 Million to Anthropic Tokens

Salesforce plans to spend approximately $300 million on Anthropic tokens in 2026, marking one of the largest commercial AI investments by a single enterprise customer

1

. Speaking on the All-In podcast, Salesforce CEO Marc Benioff disclosed that this AI spending will focus almost entirely on AI coding applications, calling both the coding agents and Anthropic "awesome"

1

3

. The investment underscores how AI integration and investment has evolved from experimental budgets into operational necessities for large technology companies.

Tokens represent the units of text that large language models process when generating output, and AI companies bill enterprise customers based on consumption volume

1

. This spending level would position Salesforce among Anthropic's largest commercial accounts. Anthropic's annualized revenue has surged from roughly $9 billion at the end of 2025 to approximately $30 billion by March 2026, driven by enterprise adoption of Claude for coding, legal, financial services, and general-purpose reasoning

1

.

Source: Benzinga

Source: Benzinga

AI Coding Agents Transform Engineer Productivity and Software Deployment

Marc Benioff emphasized that AI-driven coding functionalities have delivered significant efficiency gains across Salesforce's operations. "I can do things that I just could not do before. I can go faster than ever before. I can implement my software and sell it at the same time. I've never been able to do that before," Benioff stated on the podcast

3

. The coding use case is producing effects similar to what the company achieved in customer support, where agent-driven productivity enabled Salesforce to reduce its support workforce from 9,000 to 5,000 employees last August

1

.

Benioff stopped hiring new software engineers in 2025, opting instead to invest in AI tools that deliver better efficiency gains than adding headcount

2

. The CEO noted that productivity gains from coding agents are changing how engineers work rather than eliminating them, enabling faster product iteration, lower development costs, and a pace of output previously impossible

1

2

.

Slack Integration and Headless 360 Expand AI Capabilities

Benioff revealed that Salesforce is developing technology to enable easier coding inside Slack, the workplace communications platform Salesforce acquired for $27.7 billion in 2021

1

. "You're going to see some cool stuff with Slack and code I'm not ready to talk about yet," he teased, adding, "But there's no question that we are in a new moment in coding"

1

.

Salesforce overhauled Slack in March, unveiling more than 30 new AI capabilities that transform Slackbot into an agentic system capable of transcribing meetings, monitoring desktop activity, and executing tasks through third-party tools

1

. All new capabilities run on Claude. Slack revenue is expected to reach $3 billion this year, while Salesforce's Agentforce business hit $800 million in annual recurring revenue, up 169% year-on-year with 29,000 deals closed

1

.

Salesforce also launched Headless 360, an API-first platform with over 60 MCP tools designed to give AI agents, including Claude Code, direct access to its enterprise stack

3

.

Routing Layer Could Optimize Enterprise Market Costs

Benioff's projection includes an important caveat about cost optimization. He called for an intermediary layer that could route inputs intelligently, sending complex reasoning tasks to Claude and simpler ones to smaller, cheaper models

1

. Claude Opus 4.7, released this month, is priced at $5 per million input tokens and $25 per million output tokens. Smaller models from Anthropic's Haiku line or open-weight competitors like Meta's Llama and DeepSeek cost significantly less

1

.

At $300 million in annual spend, even modest routing optimization could save Salesforce tens of millions of dollars. Benioff appears to signal that Salesforce will build this optimization internally rather than wait for Anthropic to offer it

1

. The endorsement by the CEO of the world's largest enterprise applications vendor carries weight as token consumption becomes a major operational expense across the enterprise market.

Salesforce has invested more than $300 million in Anthropic as a company since its Series C round in early 2023, securing roughly a 1% stake in a company now valued at $380 billion

1

. Benioff called Anthropic "a rocket ship that will not stop" and noted that Microsoft blocked Salesforce from investing in OpenAI, redirecting the company toward Anthropic

1

3

. That investment has produced a paper return of more than ten times the original outlay

1

.

Today's Top Stories