The CFO AI Digest: September 4

This week, tech giants are augmenting internal capabilities through acquisitions and partnerships spanning enterprise and consumer markets. Amazon pushes AI deeper into shopping, Anthropic grows rapidly on enterprise demand, and Microsoft and Apple expand their model stacks.

Here’s what you need to know:

1. Anthropic valuation more than doubles on strength of enterprise deals

Anthropic raised $13 billion in Series F funding at a $183 billion post-money valuation, more than doubling its valuation from March. The round was co-led by ICONIQ, Fidelity Management & Research Company, and Lightspeed Venture Partners. Run-rate revenue grew from approximately $1B at the start of 2025 to over $5B by August, with Claude Code generating more than $500M in run-rate revenue since its May 2025 launch.

CFO takeaway: Enterprise adoption has driven Anthropic’s growth, Tech Funding News reports. This is consistent with the company’s reported 7x growth in large accounts over the past year. Claude Code’s impressive run rate also suggests that developer tools are becoming major revenue drivers. This is indicative of bottom-up adoption, where teams can drive early AI use before broader enterprise rollout.

Read more

2. OpenAI to acquire product experimentation platform Statsig

OpenAI plans to acquire Statsig, a product experimentation and analytics platform used for A/B testing, feature flagging, and real-time product decisioning. Statsig has already been used inside OpenAI to support fast product iteration. Following the $1.1B acquisition, Statsig founder Vijaye Raji will join OpenAI as CTO of Applications.

CFO takeaway: Statsig was built to help product and engineering teams move faster through experimentation, analytics, and release infrastructure. OpenAI’s planned purchase reflects the trend of AI companies growing product lines and making a push to reach people in new ways. Think: Anthropic’s Claude Code with Slack, or Meta's AI features showing up across WhatsApp and Instagram.

Read more

3. Amazon embeds AI deeper in shopping experience with Lens Live

Lens Live is a new AI feature within Amazon Lens, a visual search tool in the iOS Amazon Shopping app. It lets users scan items in the real world and online to find real-time product matches in a swipeable carousel in the app. The feature integrates with Rufus, Amazon’s AI assistant, to provide product summaries and answer questions.

CFO takeaway: As business-focused AI agents and chatbots move toward guiding choices and taking actions on behalf of users, consumer products are following suit. AI is becoming an embedded layer in tools people use every day, helping shape decisions through search and recommendations. This shift toward embedded decision-support could raise expectations at work, both from end users and from buyers selecting software.

Read more

4. Microsoft expands with two new in-house AI models

As part of its multi-model strategy, Microsoft is adding its own LLMs to Azure alongside GPT and open-source partner models. MAI-Voice-1 is a natural speech generation model now in Copilot Daily, Podcasts, and Copilot Labs. MAI-1-preview uses a mixture-of-experts architecture, activating only the most relevant parts of the model for each task, to deliver helpful responses that follow instructions.

CFO takeaway: Microsoft might be preparing for a future where customers expect more than just access to GPTs that can generate text and code as it expands its AI stack to include both large-scale foundation models and specialized speech generation. Similar to OpenAI’s Statsig acquisition, this suggests the tech giant is looking for ways to innovate that will help scale AI products across many surfaces.

Read more

5. Apple testing Gemini for upgraded Siri assistant

Apple is developing an AI-powered web search tool, internally dubbed World Knowledge Answers, that will debut in Siri in 2026, according to Bloomberg. As part of the effort, Apple has signed an agreement with Google to test a custom Gemini model for Siri. The updated assistant is expected to blend Apple’s in-house technology with external AI systems for web search and summarization. This places Apple in more direct competition with other leading GenAI providers.

CFO takeaway: By building its own AI search engine while also exploring partnerships with OpenAI and Google, Apple is emphasizing strategic flexibility. Even the largest, most integrated tech giants aren’t betting on a single AI model, so flexibility should be top of mind for finance leaders.

Read more

6. Federal judge puts limits on Google search exclusivity deals

In a major antitrust case ruling we covered in a previous Digest, a federal judge ruled that Google illegally maintained a monopoly in online search. While the court rejected the DOJ’s push to force a sale of Chrome or Android, it barred Google from entering into exclusive search contracts across Chrome, Google Search, Google Assistant, and the Gemini app. Google can still pursue non-exclusive default deals, but must also share portions of its search index and user-interaction data with rivals, though the data is less extensive than DOJ prosecutors had requested.

CFO takeaway: This ruling confirms what we noted when Perplexity made its bid for Chrome: a sale was unlikely, but regulatory pressure could meaningfully impact the AI landscape. By addressing exclusivity, defaults, and access to data, this decision is reshaping the conditions under which companies compete for user behavior at scale.

Read more

See you next week.

Recommended reading

1. The Second Half (Shunyu Yao)
2. The Less You Know About AI, the More You Are Likely to Use It (The Wall Street Journal). Paper referenced: Expertise, NBER Working Paper No. w33941 (National Bureau of Economic Research)

Send me your questions on AI and finance at [email protected]. We’ll pick two questions to answer each week in the following edition!