OpenAI has kicked off ambitious plans to develop an artificial intelligence agent that is aimed at replacing software engineers entirely, the company’s chief financial officer revealed this week at a major technology conference in London.
Speaking at Goldman Sachs’ Disruptive Technology Symposium, CFO Sarah Friar said the new system, known internally as A-SWE, or Agentic Software Engineer, is designed not just to assist human coders.
In fact, it will be similar to what Copilots currently do, namely to autonomously build software applications from scratch, conduct bug testing and quality assurance, and generate technical documentation.
“This is not just augmenting the current software engineers in your workforce,” Friar told the conference delegates in the British capital.
“It’s literally an agentic software engineer that can build an app for you. Not only does it build it, it does all the things software engineers hate to do,” she declared.
‘Agentic AI’ systems
A-SWE is the third in OpenAI’s growing family of “agentic AI” systems, which act independently on behalf of users.
Previous releases include Deep Research, which generates detailed research reports, and Operator, a digital assistant capable of performing complex online tasks like booking travel or making reservations.
The push to create fully autonomous AI engineers comes alongside the release of OpenAI’s latest language model, GPT-4.5, which Friar said places much more emphasis on emotional intelligence (EQ).
“We’re actually spending a bunch of time thinking about the more EQ side of models,” she explained. “We spent a lot more time training [GPT-4.5] to have what Silicon Valley loves to call ‘vibes,’ but effectively EQ.”
She described the model as “very human,” especially when applied to creative tasks like writing and design, rather than strictly technical fields like math and science.
The enhancements to GPT-4.5 come as AI adoption continues to rise, yet not without concern.
A January report by PYMNTS Intelligence found that more than half of respondents believe AI poses a significant risk to employment due to automation.
Beyond software agents, OpenAI is also undergoing a major vertical expansion, moving deeper into the data infrastructure space.
At the centre of this shift is the $500 billion Stargate project, a sprawling effort to build next-generation data centres for training and running AI systems.
The initiative, financially backed by SoftBank, Oracle, Microsoft, Nvidia and others, aims to provide OpenAI with enough computing power to avoid the bottlenecks that delayed the launch of past products, such as the video generation model Sora.
Stargate is projected to generate 10 gigawatts of power, more than the entire electricity usage of Ireland, and is key to OpenAI’s plan to become a full-stack AI provider.
“Think about Amazon at the moment they were rocking it on eCommerce,” Friar pointed out, comparing OpenAI’s move to Amazon’s launch of AWS.
“If they had decided to outsource cloud computing to Google, giving all that IP away, think about how different the company would be today.”
Friar emphasised that OpenAI is now more than a model builder. “We’re going down into data centre technology … and we feel like we’re creating a lot of IP there.”
The major engineering investment comes amid massive user growth. OpenAI now has 400 million weekly active users and has tripled revenue annually for three years running, according to Friar.
While an IPO is not imminent, she said going public is “good hygiene” and may be considered in the future. “I can’t even imagine that right now — so much going on,” she told the conference delegates.
Why not become a QA Financial subscriber?
It’s entirely FREE
* Receive our weekly newsletter every Wednesday * Get priority invitations to our Forum events *
REGULATION & COMPLIANCE
Looking for more news on regulations and compliance requirements driving developments in software quality engineering at financial firms? Visit our dedicated Regulation & Compliance page here.
WATCH NOW
READ MORE
- Antithesis swells finserv footprint as autonomous testing gains traction
- Deep Dive: will AI replace QA teams, or simply make them more valuable?
- BrowserStack snaps up Requestly to expand open-source testing
- Eximius leads pre-seed round in rising QA star DevAssure
- AI testing takes centre stage in Bank of England’s data transformation
