OpenAI Considering $20,000/mo Tier for "PhD AI Researcher" - It May Be a Bargain
Agentic AI application like Open AI Deep Research at $20/mo are already transformational
The Information reports a potential new pricing tier for OpenAI at $20,000/month.
A $20,000/month tier is a shrewd move by OpenAI to test the "outsources knowledge worker" market. These are AI tools that increase the productivity of an individual by 2x to 30x for some (not all) tasks.
As you likely know, I'm very impressed with the entire category of agentic AI applications for research that are available now and that are already radically changing work. Users type their research query into the prompt or the "box" and then the AI application "thinks" or deploys individual AI agents to specific tasks like clarifying user intent, searching internet for public sources, evaluating sources, and producing a report. After 6-15 minutes, a 10-15 page research report is produced. Amazing. I've done tests in dozens of areas with subject matter experts, and they're impressed.
Products in this "AI Research Agentic Application" category include OpenAI Deep Research, DeepSeek V3, Grok Deep Research, Perplexity's Deep Research and others.
Per The Information article, OpenAI is considering the following agent pricing:
$2,000 per month to "high-income knowledge workers"
Mid-tier agents for software development costing $10,000 a month
PhD-level research agents at $20,000 per month
For some companies, this will be a bargain. The notion is not full employee replacement, but smaller team of human experts augmented by AI. AI is simply superpowers for knowledge workers. Here is an example:
An explosive and historic change has already started for knowledge workers.
We are now in the adapt stage. For adaptive businesses, this means AI-augmentation for small teams.
What are your thoughts?
MIT, John Werner, and Imagination in Action continue to host amazing AI events.
The next one is April 15 at MIT. Tickets are free, but you must apply here.
This is a fantastic opportunity for you and your team to learn, to network, and to accelerate your path to driving. Here is video highlights from this conference last year.
Let me know if you are attending.
Check out these amazing, confirmed speakers.
Sridhar Ramaswamy, CEO, Snowflake
Stephen Wolfram, Founder and CEO, Wolfram Research
Andrew Ng, Founder, DeepLearning.AI
Douwe Kiela, CEO and Co Founder, Contextual AI
Chase Lochmiller, CEO and Co-Founder, Crusoe
Jim Rowan, Head of AI, Deloitte
Paul Baier, Co-Founder and CEO, GAI Insights
John Sviokla, Chairman, Executive Fellow, GAI Insights
Emrecan Dogan, Head of Product, Glean
Peter Grabowski, Lead, Gemni Research, Google Core ML
Karim Lakhani, Professor, Harvard
Jean Olive, CIO, John Hancock
Geoffrey von Maltzahn, , CEO, Lila AI
Ramin Hasani, Co-founder and CEO, Liquid AI
Amit Jain, CEO, Luma
Navin Chaddha, Managing Partner, Mayfield
Mark Gorenberg, Chair, MIT Corporation
Daniela Rus, CSAIL Director, MIT CSAIL
Ramesh Raskar, Professor, MIT Media Lab
Pattie Maes, Professor, MIT Media Lab
David Kenny, Executive Chairman, Nielsen
Jane Urban, Vice President, Customer Engagement, Otsuka Pharmaceutical
Steve Papa, Founder and CEO, Parallel Wireless
Rodrigo Liang, , Co-founder and CEO, SambaNova Systems
Jack Hidary, CEO, Sandbox AI
Prem Akkaraju, CEO, Stability AI
Sandy Pentland, Senior Research Fellow, Stanford Digital Economy Lab
Anurang Revri, VP, Chief Enterprise Architect, Stanford Healthcare
Cindi Howson, Chief Data and AI Strategy Officer, ThoughtSpo
Dan Slagen, Chief Marketing Officer, Tomorrow.io
Ulrike Hoffmann-Burchardi, Chief Investment Officer, UBS Global Wealth
Karen Beck Stroup, Chief Digitial Officer, WEX
I’ve been raving about OpenAI Deep Research product for weeks — put in a prompt, wait 10 minutes and get a 15 page research paper will full citations.
My previous post, We Monkey-Brain Humans Must Adapt to an AI World. The Tools are Smarter Than Us, generated some good email discussion. I argue that our AI tools are smarter than us most cognitive knowledge work. We need to expect massive impact on many >$100K jobs and get serious about adaption.
What are your thoughts?
Join us Fri, March 14 at Noon with executives from 3 major health insurance companies: BCBS of Michigan, Horizon BCBS of New Jersey and Cambia Health Solutions. Sign up here
Many companies tell us that it is important for them to offer a Secure Employee ChatBot this year for 10-30% of their employees to improve productivity and employee satisfaction. To reduce confusion, our AI Analysts compiled this feature comparison guide.
Goldman Sachs now has Secure Employee ChatBot to 20% of its 50,000 employees and aims to have 100% coverage by year end.
Apply to speak, exhibit and pitch your AI Startup at 3rd Annual conference, GAI World 2025, celebrating and learning from corporate leaders IN PRODUCTION with GenAI corporate applications. Our conference is Sep 29-30 in Boston (and we sold out the last 2 years). More info here
“A distracted mind is a weak mind” — Bert McCoy
Onward,
Paul
P.S. You are one of 9,300 subscribers to this newsletter. Please share with a colleague.