Salesforce to Spend $300 Million on Anthropic AI Tokens Salesforce is dramatically escalating its AI bet, planning to spend $300 million on Anthropic tokens in 2026 as it shifts software development toward an AI-heavy, “supervisory” model for human engineers.

Early signals: productivity surge and hiring freeze

Salesforce’s pivot began in 2024, when CEO Marc Benioff told investors that AI tools like Agentforce had boosted engineering productivity by more than 30%, prompting a halt to new software engineering hires. Instead, the company expanded hiring in sales and other roles to explain and sell its AI products.

By 2025, Benioff was describing a “new labour model,” with roughly 15,000 engineers working in a supervisory capacity over AI systems such as Anthropic’s Claude, OpenAI tools, and Cursor. He argued that AI was transforming engineering work rather than replacing developers outright.

The $300 million Anthropic token plan

On the All-In podcast in May 2026, Benioff said Salesforce expects to spend $300 million on Anthropic tokens this year, “almost entirely on coding,” calling AI coding agents and Anthropic “awesome” and predicting that the outlay would make “everything at Salesforce cheaper to build.” Industry reports framed this as Salesforce planning a US$300m spend on Anthropic’s AI tokens in 2026, primarily for coding applications to lower software development costs and increase output.

Analysts note that a $300 million annual token bill would make Salesforce one of Anthropic’s largest commercial customers, amid Anthropic’s rapid revenue growth driven by enterprise use of Claude for coding, legal, financial, and general-purpose reasoning tasks.

Slack as the next AI coding hub

Benioff has also linked the spending to a broader product vision. He has said Salesforce is working on tech to “make coding easier inside Slack,” the $27.7 billion collaboration platform the company acquired in 2021. In March, Salesforce overhauled Slack with more than 30 new AI capabilities powered by Anthropic’s Claude, turning Slackbot into an agentic system that can transcribe meetings, monitor activity, execute tasks through third-party tools, and act as a lightweight CRM.

Across these moves, Benioff is also advocating for an intermediary “routing layer” that can direct workloads between frontier and smaller models to balance cost and performance, underscoring Salesforce’s view that AI tokens are becoming a core input to digital labor rather than just another software expense.

Continue reading https://foxvector.com