OpenAI announces UAE deployments as Altman describes AI gains
OpenAI’s Altman said advanced AI models had become mostly smarter than the brightest scientists, and said AI could perform significant parts of many white-collar jobs. OpenAI also announced partnerships with G42 and ADNOC during the event.

Key takeaways · 4
- 01
Altman said advanced AI models had become mostly smarter than the brightest scientists.
- 02
ADNOC, G42 and OpenAI agreed to deploy OpenAI frontier technology across ADNOC operations.
- 03
ADNOC said its system would use GPT models on infrastructure inside the UAE, with experts overseeing consequential decisions.
- 04
OpenAI, Mubadala and Inception42 agreed to apply advanced AI capabilities to investment analysis.
Capability claims and human control
Altman said AI had shifted from being mostly less capable than the smartest scientists at the start of the year to mostly smarter than them.[1] He also said AI could now perform significant parts of many white-collar jobs.[1] Alongside those claims, he said people must remain in control of AI’s role in shaping the future and rejected handing decisions over to AI.[1] He described AI as a tool that should serve people.[1] For teams assessing AI use, the remarks pair a claim of expanding capability with an explicit emphasis on human control.[1]
Safety, scale and energy
Altman asked how to guarantee avoidance of serious safety and alignment incidents and give the world comfort.[2] He identified energy as the challenge he faced that day.[2] He also said the physical world powering AI would dominate its cost and abundance.[1] On usage, he said global per-capita token use had risen from near zero six years earlier to about 100,000 tokens a month, while top institutional users consumed about 500 billion tokens monthly.[1] Those figures put infrastructure and energy alongside safety as issues raised in his comments.[2][1]
ADNOC deployment details
ADNOC, G42 and OpenAI agreed to deploy OpenAI frontier technology across ADNOC’s operations.[3] ADNOC said the system would use OpenAI GPT models on infrastructure inside the UAE for field planning and to connect operational, capital and commercial decisions.[2] The company said ADNOC experts would oversee consequential decisions.[2] That description specifies a deployment setting and intended work areas, while retaining expert oversight for consequential decisions.[2] It does not describe a transfer of decision-making authority to the system, consistent with Altman’s stated position that people should remain in control.[1][2]
Other UAE initiatives
OpenAI, Mubadala and Inception42 agreed to apply advanced AI capabilities to investment analysis.[3] The Abu Dhabi Government announced an initiative with an AI lab to translate government policies and regulations into machine-readable form.[3] Separately, Inception42 and OpenAI signed a five-year agreement to support adoption of frontier AI across government, enterprise and education.[3] OpenAI was also seeking to raise $20 billion in a new funding round co-led by UAE investors.[3] The announcements span operational, investment and public-sector uses, as well as a funding effort.[3]
The announcements offer professionals concrete examples of AI being positioned for operational planning, investment analysis and government work. Teams evaluating similar uses can pay attention to the stated deployment conditions, expert oversight and the energy and safety concerns Altman raised.
Why it matters
Bring this to your team — one session a day, built for your industry.
kju.ai is built for teams: daily sessions, team analytics, and rollout support for your organization.
For teamsHow this developed
2 November 2026
OpenAI announces UAE deployments as Altman describes AI gains