What will $500 billion poured into AI infrastructure over the next four years in the United States accomplish? CIOs and CTOs will have to watch the Stargate Project to find out. The initiative -- a collaboration between several high-profile players in the AI space -- has been plugged by President Trump. Billions are already being invested, and construction on several data centers has already begun, AP news reports.  With competition for AI dominance at a fever pitch, how much of a role could Stargate have in tipping the scales in favor of the US?  Stargate Partners  Stargate has a lot of AI star power behind it.  “The players that are part of the project are all people who are very invested in building out a computing infrastructure for AI already and building frontier AI systems,” says Peter N. Salib, law and policy advisor to the nonprofit Center for AI Safety and codirector of the Center for Law & AI Risk, an organization focused on establishing law and AI safety as a scholarly field.  OpenAI, software company Oracle, AI investment firm MGX, and investment holding company SoftBank are the four initial equity funders powering the project. SoftBank is taking the lead on financial responsibility, while OpenAI is tackling operations.  Related:The project has also attracted technology partners, including semiconductor company Arm, tech giant Microsoft, and chip company Nvidia  That is a lot of cooks in the kitchen, all very motivated to push the field of AI forward, ultimately achieving AGI.  “What I hope is that this becomes a model or an example of how titans of industry and government and ultimately and eventually the community are able to work together for the benefit of mankind,” says Jason Hardy, CTO for AI at data infrastructure company Hitachi Vantara. Time will tell if each partner delivers on their promises and ultimately plays well with others. “I would effectively call this a moonshot. So, it'll be interesting to see over the next year or so how it progresses,” Hardy adds.  The Goals The Stargate Project is focused on the “development and construction of large-scale AI data centers,” according to its request for proposals. And there is no doubt about AI’s voracious appetites. Advances in the field -- feeding those appetites -- will require more infrastructure, but is that alone the answer to capturing the lead? Randall Hunt, CTO at Caylent, an AWS cloud consulting and engineering company, thinks not. “Infrastructure alone is a very brute force approach to solving artificial general intelligence or artificial super intelligence,” he says.  Related:He goes on to voice some additional goals that could be beneficial in the pursuit of AI advancements.  “I think that we need significant efficiency and architectural improvements in the underlying implementation of these networks,” Hunt argues. “And I think a broader initiative that focuses not just on pure infrastructure but also investment in theoretical work and investment in academia and investment and the software side would be pretty valuable.” Funding In January, Elon Musk took to X, claiming Stargate lacked the funding. OpenAI’s CEO Sam Altman refuted the claim, and a source told Forbes that the initial $100 billion in equity is “ready to go.” That still leaves $400 billion to be gathered over the next four years.  “Sometimes … in these large agreements, you'll have pledges of capital and then you have investor underperformance. They don't ever actually send the desired capital,” Hunt points out.  Stargate is in its early days, and while a shortfall in funding for this project is possible, overall spending on AI and its requisite infrastructure is likely to soar well beyond the $500 billion point in the coming years.  “I will never underestimate the private sector’s ability and desire to put up money towards this race,” says Caroline Winnett, executive director of startup accelerator Berkeley SkyDeck.  Related:Apple, for example, is pumping $500 billion into US facilities, including a server production facility for its AI products, CNN reports.  Performance Metrics The initiative already has data centers under construction in Texas and several other states lined up as possibilities for its campuses, Reuters reports. How will we know if Stargate is delivering on its goals and putting those billions to good use?  At the most basic level, we can look at data center capacity. How many megawatts have resulted from the buildout of Stargate’s data centers? There are, of course, more nuanced questions about efficiency and energy usage.  “Does putting more data through more compute continue to get you ever more capable systems? All the … evidence we have seems to point towards the prediction that it will, and if that turns out not to be true, that will be a big surprise and a big knock against this Stargate approach of doing extremely large-scale compute and power clusters,” says Salib. The ultimate goal of AGI looms large. Will Stargate and its participants be the first to achieve it? Tracking the Global AI Race The emergence of DeepSeek from a Chinese startup threw fuel on the competitive fires. And OpenAI is certainly cognizant of the flames. "As news emerged about DeepSeek, it makes it clear this is a very real competition and the stakes could not be bigger,” said Chris Lehane, OpenAI’s chief global affairs officer, Reuters reports.  Among the slew of executive orders Trump signed upon taking offices is one aimed at American leadership in AI.  “I think it's true to say that the US and China do understand themselves as racing towards something like artificial general intelligence and that this project might make help them to race faster,” says Salib.  Stargate could propel the US forward in this breakneck sprint, but it needn’t be the initiative on which the country pins all of its hopes. Moonshots like this are not guaranteed successes. But it is hardly a solo shot.  “Stargate is emblematic of the scale of which frontier AI is going to be developed in the next two to five years, but it's not the only project that is going to look the way that Stargate looks,” says Salib.  Even if Stargate fizzles out for one reason or another, it is highly unlikely that the US will find itself falling completely behind. There will still likely be plenty to learn from the endeavor, and there will be other projects and players with skin in the game.  “Whether this initiative moves forward or not, no matter what happens with it, everybody's going after this golden carrot known as AGI,” Winnett points out.  Predictions on the arrival of AGI vary, but it seems all but certain that it is coming. And the road there is hardly written in the stars. There is still plenty of room for surprises and disruption.  “People think these entrenched players like OpenAI and Anthropic and AWS, that they've got a moat that can't be overcome, but we're still in the wild west days,” says Hunt. “The model that's winning today is not necessarily the model that's winning tomorrow. As tech companies and governments pound the pavement in this ongoing race, there are some big, open questions.  “A lot of regulation is going to need to be looked at and evaluated to see how we can improve on power generation. Does nuclear need to be a part of it, for example?” says Hardy.  And then, there are other thorny concepts to grapple with: What is the cost of racing in the first place? Is the world ready for what it means when we reach a point where a winner can be declared?  “As with say the missile gap of the Cold War era, racing has its own dangers,” Salib points out. “Both sides would really like to have the most powerful systems as quickly as possible and seem willing to risk losing control of their own systems for the sake of winning that capabilities race.” Along the way, the environmental strain and energy usage associated with AI has costs.  “We're hoping that AI can produce solutions that will actually make very significant progress [on] how these tools end up interacting with the environment [and] solve their own issues,” says Winnett.  For now, it seems that the race is still on, whether or not those solutions materialize.  As AGI grows closers, Salib hopes we will spend more time thinking not only about its value but its risks. “The risks of misuse of these extremely powerful systems, arms races around these extremely powerful systems, and also loss of control the systems themselves as they become very capable. It is time for all of us to take all of that very seriously in a way that I think most of the policy world is not yet,” he urges.