This week in AI: OpenAI’s ‘Operator’; US invests $100B in data center

Getting your Trinity Audio player ready...

OpenAI launches AI agent ‘Operator’

On January 23, OpenAI unveiled “Operator,” its AI agent designed to execute complex, multi-step tasks autonomously. In a live demonstration, Operator handled tasks such as making restaurant reservations, ordering groceries, and purchasing event tickets from start to finish, with minimal user interaction beyond entering the initial prompt.

Operator runs on OpenAI’s Computer Using Agent Model (CUA), built on GPT-4o. This system allows artificial intelligence (AI) to open a cloud-based browser and execute tasks with user-defined guidelines while giving users the ability to intervene at any stage to override and take control from the autonomous system.

Currently, Operator is only available to GPT Pro users in the U.S.—those who pay $200 per month for OpenAI’s services.

The launch marks a significant moment in AI. For several months, other companies have previewed and privately tested their agents but have ultimately not delivered a public AI agent like OpenAI has. In addition, many of these early AI agents operated within their own walled gardens and had limited capabilities. However, OpenAI’s Operator, with its ability to handle complex, multi-step tasks, seems closer to fulfilling the original vision of agents the world was promised: an AI system that can independently manage multi-step tasks from start to finish.

The $100 billion AI data center initiative in the US

On January 21, President Donald Trump, Oracle CEO Larry Ellison, SoftBank (NASDAQ: SFTBY) CEO Masayoshi Son, and OpenAI CEO Sam Altman announced “Stargate,” an artificial intelligence data center venture. Stargate is a joint effort led by Oracle, SoftBank, and OpenAI, aiming to invest a minimum of $100 billion in developing new AI-driven data centers across the United States.

During their press conference, the team stated that 10 500,000-square-foot data centers are already under construction in Texas and that they hope to quickly increase that number to 20 data centers. In addition to the $100 billion investment, the group announced plans to raise up to $500 billion to expand the project, with contributions expected from other major tech companies.

Shortly after the announcement, Elon Musk criticized the venture, claiming it did not have the $100 billion in funding it asserted. However, Microsoft (NASDAQ: MSFT) CEO Satya Nadella quickly countered, announcing Microsoft’s commitment to invest $80 billion in Stargate on top of the money already invested by the three companies leading the initiative.

Microsoft $MSFT CEO Satya Nadella, commenting on Elon Musk's funding concerns over the $500 billion AI infrastructure project involving OpenAI, Oracle and SoftBank, Stargate, said, “All I know is, I’m good for my $80 Billion.”
pic.twitter.com/kWCsf9KutB

— Asia Tech (@asiatechwire) January 23, 2025

The StarGate project comes at a crucial moment in the artificial intelligence industry. The surge in AI products and services over the last three years has put a strain on data centers, which provide the computing power and infrastructure required for AI operations. Without expanding and building new data centers, AI development risks being stifled due to the capacity constraints of existing facilities.

The tech giants creating AI goods and services have been vocal about this issue and privately invested in building and expanding clean-energy-powered data centers. But now, the government is becoming involved. Toward the end of Biden’s administration, efforts began to lease federal land for clean-energy-powered AI data centers. Although Trump repealed nearly all of Biden’s executive orders upon taking office, his administration is carrying forward plans to expand data center infrastructure and clean energy in America.

LinkedIn accused of misusing private messages to train AI

LinkedIn faces a class-action lawsuit in the United States, alleging that the platform used private user messages to train its AI models without proper consent. The lawsuit claims LinkedIn shared this data with third parties and subsequently updated its privacy policy to make it seem as though such practices were always permissible.

LinkedIn has denied these allegations, calling the claims baseless. However, the lawsuit highlights a broader concern with generative AI: data security and transparency. Many users unknowingly grant AI platforms access to their data through terms and conditions. Even if a user knowingly gives the AI platform access and approval to use their data, it remains unclear how the platform uses that data and whether it’s shared with third parties. These issues may occur but are not clearly addressed in the terms of service agreements the AI platform has with its end users.

When dealing with sensitive information, this poses significant risks. For example, entering credit card details into an AI chatbot could inadvertently lead to that data appearing in future outputs. This is just one example, but it’s one of the reasons why corporations caution—and sometimes outright ban—their employees from using generative AI or entering sensitive information into these systems.

While it’s unclear whether LinkedIn is guilty of the accusations, the case serves as a reminder for users to be cautious when interacting with generative AI tools. Without explicit guarantees of data security, users should assume their inputs might be retained or used in ways they didn’t anticipate.

In order for artificial intelligence (AI) to work right within the law and thrive in the face of growing challenges, it needs to integrate an enterprise blockchain system that ensures data input quality and ownership—allowing it to keep data safe while also guaranteeing the immutability of data. Check out CoinGeek’s coverage on this emerging tech to learn more why Enterprise blockchain will be the backbone of AI.

Watch: Turning AI into ROI