Your new $240,000 Digital Colleague
OpenAI replaces your team with Digital Workers
Sharing real-world enterprise AI use cases, straight from conversations with early adopters actually deploying AI and enterprise agents. No fluff. Just what’s working, what’s not, and what’s next in AI & Agents for business. 🍺
It was reported this week that OpenAI is exploring an update to their Agent pricing, to the tune of $240,000 annually 🤯
According to the Information, OpenAI is reportedly planning to release multiple AI Agent products tailored for different applications, such as Sales use cases (look out AI SDRS!) and software development. There is a “high-income knowledge worker” Agent expected to cost $2,000/month, while a software developer Agent is said to be priced at $10,000/month.
The most expensive rumored offering, priced at $20,000 per month—$240,000 a year— is designed to support PhD-level research.
To date, OpenAI has largely operated as a consumer product, and a successful one at that—400 million WAU, the #1 app on the App Store, and a $20/month pricing model (for context, Netflix has ~300m subscribers).
However, in 2024, the company reportedly lost $5 billion on $3.7 billion in revenue. Forecasts suggest losses could rise to $14 billion by 2026, with cumulative losses reaching $44 billion between 2023 and 2028. Profitability isn’t expected until 2029, when revenue is projected to hit $100 billion.
At some point they have to start making money. Right?
Today, OpenAI segments their sellers by 1) Chat and 2) API.
Chat remains the workhorse for both Consumer and Enterprise, which they sell on a basic seat model. I spoke to an Enterprise recently that was planning to purchase GPT seats for 3k employees. That’s a quick $720k annual agreement with OpenAI (don’t do this btw! I have a better approach.)
Side note: Seeing high adoption of Google Gemini. Security teams seem comfortable enabling Gemini as part of their Google Workspace. Same goes for Microsoft Copilot. My bet would be on these two winning the generic “chat” use case in the Enterprise.
They’re also scaling their Enterprise efforts and have closed some absolutely massive ($1b+) deals to help organizations build on OpenAI’s APIs, which are priced based on token utilization. For example, they’re working with a leading Telco to build Agents to manage all S1 support interactions, starting with chat and expanding to voice. They’re expecting to process millions of support requests annually. This carries a huge ROI with clear metrics such as case deflection, average handle time, CSAT, and (Altman’s favorite) human replacement.
Enter packaged Agents. It seems the plan is to anchor pricing relative to the job(s) they believe the Agent can replace. If pricing truly mirrors job replacement value, Enterprises might weigh the cost of multiple Agents against full-time hires. Will companies pay per Agent, per task, or will OpenAI shift to a usage-based model that reflects output rather than headcount? Hopefully, they forget that Agents aren’t bound to 40-hour work weeks.
We live in a fast-follow world right now. What will prevent Anthropic, Cohere, X, etc.. to just quickly undercut this pricing with a similar (if not better) model?
Based on Altman’s recent tweet, it seems like OpenAI has no idea who to hell to price either, so who knows..
Until next week, when everything changes. 🤷♂️
🔗 Dive deeper on this topic
[Substack] - best overview I’ve found on AI pricing models from
:
🤖 Enterprise AI use case of the week
Use case: This Agent is a Pardot expert out of the box and can help users create prospects, find prospects, update prospects, create and delete custom fields, create lists, add prospects to lists, create campaigns, and list custom fields all from a simple chat interface.
Who is this for: Marketing, Sales, GTM
Tools used: Pardot, Tray AI
I started this newsletter because I am frustrated by the lack of tangible Enterprise AI use cases in the market.
I have the opportunity to speak to hundreds of tech and systems leaders and fundamentally believe that AI and Agents will change the way businesses operate. My goal is to help share how.
-Nate G
PS - I went to Regionals in “Power of the Pen” in 8th grade so I consider myself a fairly prolific writer.