OpenAI cozies up to Uncle Sam with ChatGPT government edition
(2025/01/28)
- Reference: 1738094407
- News link: https://www.theregister.co.uk/2025/01/28/openai_us_government/
- Source link:
OpenAI has announced ChatGPT Gov, a variant of the Enterprise version of the product specifically tailored for use by the US government.
Coinciding with revelations that Chinese AI firm DeepSeek's R1 model may [1]undermine the hard work done on US-made AI, ChatGPT Gov was [2]announced today as a way to, according to OpenAI, ensure President Trump's executive [3]order on removing barriers to American AI leadership succeeds.
"By making our products available to the US government, we aim to ensure AI serves the national interest and the public good, aligned with democratic values, while empowering policymakers to responsibly integrate these capabilities to deliver better services to the American people," OpenAI said in its announcement.
[4]
ChatGPT Gov is more about security than new features, with most of the changes being made to ensure ChatGPT conforms to US government standards.
[5]
[6]
According to OpenAI, ChatGPT Gov can be deployed on top of Microsoft's Azure OpenAI Service and can be run in the Azure commercial or Government clouds, as well as being self-hosted to "enable agencies to more easily manage their own security, privacy, and compliance requirements."
The AI giant said it also hopes ChatGPT Gov will help "expedite internal authorization of OpenAI's tools for the handling of non-public sensitive data," which doesn't appear to be the case yet. While Azure OpenAI and GPT 4o have been [7]granted FedRAMP authorization, which is the US government's Federal Risk and Authorization Management Program, ChatGPT has yet to be granted approval to handle sensitive, non-public data.
[8]
The Microsoft-backed firm said it's also looking at getting ChatGPT Gov expanded for use in [9]Azure classified regions, though it didn't say when that expansion may actually occur.
ChatGPT Gov won't be OpenAI's first foray into US government operations. In January 2024 - days after revising its policy language to [10]remove explicit prohibitions on military and warfare applications, among other things - the biz announced plans to develop [11]AI-powered cybersecurity capabilities for the Pentagon.
[12]OpenAI wants to blow through $500B on AI infrastructure for itself, with help from pals
[13]DeepSeek's R1 curiously tells El Reg reader: 'My guidelines are set by OpenAI'
[14]Microsoft injects ChatGPT into 'secure' US government Azure cloud
[15]Microsoft is a national security threat, says ex-White House cyber policy director
OpenAI pointed out in its ChatGPT Gov announcement that it's also been in use at the Air Force Research Laboratory, where ChatGPT Enterprise is being used for administrative work, at Los Alamos National Laboratory, where Enterprise is being used "for scientific research and innovation," and by the State of Minnesota for faster language translation.
The outfit also boasts its [16]pilot program in Pennsylvania , which began in early 2024, has been a rollicking success. The "first-in-the-nation AI pilot program found ChatGPT Enterprise helped reduce the time spent on routine tasks … by approximately 105 minutes per day on the days they used it," OpenAI said.
We reached out to OpenAI to get some context for that 105 minute figure, and ask additional questions about the capabilities of ChatGPT Gov, but didn't hear back.
[17]
"Since 2024, more than 90,000 users across more than 3,500 US federal, state, and local government agencies have sent over 18 million messages on ChatGPT," OpenAI said. "We look forward to collaborating with government agencies to enhance service delivery to the American people through AI and to foster public trust in this critical technology." ®
Get our [18]Tech Resources
[1] https://www.theregister.com/2025/01/27/tech_stocks_tank_as_us/
[2] https://openai.com/global-affairs/introducing-chatgpt-gov/
[3] https://www.whitehouse.gov/presidential-actions/2025/01/removing-barriers-to-american-leadership-in-artificial-intelligence/
[4] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/aiml&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=2&c=2Z5lhjiqfLBQIO550D_9WWQAAAQs&t=ct%3Dns%26unitnum%3D2%26raptor%3Dcondor%26pos%3Dtop%26test%3D0
[5] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/aiml&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=4&c=44Z5lhjiqfLBQIO550D_9WWQAAAQs&t=ct%3Dns%26unitnum%3D4%26raptor%3Dfalcon%26pos%3Dmid%26test%3D0
[6] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/aiml&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=3&c=33Z5lhjiqfLBQIO550D_9WWQAAAQs&t=ct%3Dns%26unitnum%3D3%26raptor%3Deagle%26pos%3Dmid%26test%3D0
[7] https://devblogs.microsoft.com/azuregov/azure-openai-fedramp-high-for-government/
[8] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/aiml&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=4&c=44Z5lhjiqfLBQIO550D_9WWQAAAQs&t=ct%3Dns%26unitnum%3D4%26raptor%3Dfalcon%26pos%3Dmid%26test%3D0
[9] https://www.theregister.com/2023/11/20/microsoft_azure_government_cloud/
[10] https://www.theregister.com/2024/01/15/ai_in_brief/
[11] https://www.theregister.com/2024/01/16/us_military_openai/
[12] https://www.theregister.com/2025/01/22/openai_stargate_ai_datacenter_company/
[13] https://www.theregister.com/2025/01/27/deepseek_r1_identity/
[14] https://www.theregister.com/2023/06/08/microsoft_chatgpt_us_government/
[15] https://www.theregister.com/2024/04/21/microsoft_national_security_risk/
[16] https://www.theregister.com/2024/01/11/pennsylvania_openai_chatgpt/
[17] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/aiml&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=3&c=33Z5lhjiqfLBQIO550D_9WWQAAAQs&t=ct%3Dns%26unitnum%3D3%26raptor%3Deagle%26pos%3Dmid%26test%3D0
[18] https://whitepapers.theregister.com/
Coinciding with revelations that Chinese AI firm DeepSeek's R1 model may [1]undermine the hard work done on US-made AI, ChatGPT Gov was [2]announced today as a way to, according to OpenAI, ensure President Trump's executive [3]order on removing barriers to American AI leadership succeeds.
"By making our products available to the US government, we aim to ensure AI serves the national interest and the public good, aligned with democratic values, while empowering policymakers to responsibly integrate these capabilities to deliver better services to the American people," OpenAI said in its announcement.
[4]
ChatGPT Gov is more about security than new features, with most of the changes being made to ensure ChatGPT conforms to US government standards.
[5]
[6]
According to OpenAI, ChatGPT Gov can be deployed on top of Microsoft's Azure OpenAI Service and can be run in the Azure commercial or Government clouds, as well as being self-hosted to "enable agencies to more easily manage their own security, privacy, and compliance requirements."
The AI giant said it also hopes ChatGPT Gov will help "expedite internal authorization of OpenAI's tools for the handling of non-public sensitive data," which doesn't appear to be the case yet. While Azure OpenAI and GPT 4o have been [7]granted FedRAMP authorization, which is the US government's Federal Risk and Authorization Management Program, ChatGPT has yet to be granted approval to handle sensitive, non-public data.
[8]
The Microsoft-backed firm said it's also looking at getting ChatGPT Gov expanded for use in [9]Azure classified regions, though it didn't say when that expansion may actually occur.
ChatGPT Gov won't be OpenAI's first foray into US government operations. In January 2024 - days after revising its policy language to [10]remove explicit prohibitions on military and warfare applications, among other things - the biz announced plans to develop [11]AI-powered cybersecurity capabilities for the Pentagon.
[12]OpenAI wants to blow through $500B on AI infrastructure for itself, with help from pals
[13]DeepSeek's R1 curiously tells El Reg reader: 'My guidelines are set by OpenAI'
[14]Microsoft injects ChatGPT into 'secure' US government Azure cloud
[15]Microsoft is a national security threat, says ex-White House cyber policy director
OpenAI pointed out in its ChatGPT Gov announcement that it's also been in use at the Air Force Research Laboratory, where ChatGPT Enterprise is being used for administrative work, at Los Alamos National Laboratory, where Enterprise is being used "for scientific research and innovation," and by the State of Minnesota for faster language translation.
The outfit also boasts its [16]pilot program in Pennsylvania , which began in early 2024, has been a rollicking success. The "first-in-the-nation AI pilot program found ChatGPT Enterprise helped reduce the time spent on routine tasks … by approximately 105 minutes per day on the days they used it," OpenAI said.
We reached out to OpenAI to get some context for that 105 minute figure, and ask additional questions about the capabilities of ChatGPT Gov, but didn't hear back.
[17]
"Since 2024, more than 90,000 users across more than 3,500 US federal, state, and local government agencies have sent over 18 million messages on ChatGPT," OpenAI said. "We look forward to collaborating with government agencies to enhance service delivery to the American people through AI and to foster public trust in this critical technology." ®
Get our [18]Tech Resources
[1] https://www.theregister.com/2025/01/27/tech_stocks_tank_as_us/
[2] https://openai.com/global-affairs/introducing-chatgpt-gov/
[3] https://www.whitehouse.gov/presidential-actions/2025/01/removing-barriers-to-american-leadership-in-artificial-intelligence/
[4] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/aiml&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=2&c=2Z5lhjiqfLBQIO550D_9WWQAAAQs&t=ct%3Dns%26unitnum%3D2%26raptor%3Dcondor%26pos%3Dtop%26test%3D0
[5] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/aiml&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=4&c=44Z5lhjiqfLBQIO550D_9WWQAAAQs&t=ct%3Dns%26unitnum%3D4%26raptor%3Dfalcon%26pos%3Dmid%26test%3D0
[6] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/aiml&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=3&c=33Z5lhjiqfLBQIO550D_9WWQAAAQs&t=ct%3Dns%26unitnum%3D3%26raptor%3Deagle%26pos%3Dmid%26test%3D0
[7] https://devblogs.microsoft.com/azuregov/azure-openai-fedramp-high-for-government/
[8] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/aiml&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=4&c=44Z5lhjiqfLBQIO550D_9WWQAAAQs&t=ct%3Dns%26unitnum%3D4%26raptor%3Dfalcon%26pos%3Dmid%26test%3D0
[9] https://www.theregister.com/2023/11/20/microsoft_azure_government_cloud/
[10] https://www.theregister.com/2024/01/15/ai_in_brief/
[11] https://www.theregister.com/2024/01/16/us_military_openai/
[12] https://www.theregister.com/2025/01/22/openai_stargate_ai_datacenter_company/
[13] https://www.theregister.com/2025/01/27/deepseek_r1_identity/
[14] https://www.theregister.com/2023/06/08/microsoft_chatgpt_us_government/
[15] https://www.theregister.com/2024/04/21/microsoft_national_security_risk/
[16] https://www.theregister.com/2024/01/11/pennsylvania_openai_chatgpt/
[17] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/aiml&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=3&c=33Z5lhjiqfLBQIO550D_9WWQAAAQs&t=ct%3Dns%26unitnum%3D3%26raptor%3Deagle%26pos%3Dmid%26test%3D0
[18] https://whitepapers.theregister.com/
abend0c4
I can't help feeling that they're a late arrival at the hallucinatory government ball.
we aim to ensure AI serves the national interest and the public good, aligned with democratic values
Howard Sway
Translation : After making a $1 million donation to the new boss, we've mysteriously been awarded a government contract, which is pretty fortunate because now that that free Chinese thing has eaten our lunch in the marketplace, we can still survive off a steady steam of taxpayers money.
ChatGPT Gov was announced today as a way to,...
EricM
... according to OpenAI, ensure President Trump's executive order on removing barriers to American AI leadership succeeds.
I always assumed Americans do not have a sense for Irony.
I stand corrected...
got every single question wrong
David Newall
A colleague browbeat me into trying o1 on a web development project. Not a single one of its answers was correct, but it did give the kernel of a useful solution which, through additional reading and experimentation, resulted in a very good solution. I'm sure it would have taken me much longer to get there without o1.
The takeaway for me is that it one treats everything it says with utmost suspicion and checks it carefully, it most certainly can help an experienced practitioner arrive at good solutions. Take nothing it says at face value or you'll end up looking like a fool; which would be well deserved.
"conforms to US government standards"
Bullcrap.