As winter descended on San Francisco in late 2022, OpenAI quietly pushed a fresh service dubbed ChatGPT live with a blog post and a single tweet from CEO Sam Altman. The squad labeled it a “low-key investigation preview” — they had good reason to set expectations low.
“It couldn’t even do arithmetic,” Liam Fedus, OpenAI’s head of post-training says. It was besides prone to hallucinating or making things up, adds Christina Kim, a investigator on the mid-training team.
Ultimately, ChatGPT would become anything but low-key.
While the OpenAI researchers slept, users in Japan flooded ChatGPT’s servers, crashing the site only hours after launch. That was just the beginning.
“The dashboards at that time were just always red,” recalls Kim. The launch coincided with NeurIPS, the world’s premier AI conference, and shortly ChatGPT was the only thing anyone there could talk about. ChatGPT’s mistake page — “ChatGPT is at capacity right now” — would become a acquainted sight.
“We had the first launch gathering in this tiny room, and it wasn’t like the planet just lit on fire all of a sudden,” Fedus says during a fresh interview from OpenAI’s headquarters. “We’re like, ‘Okay, cool. I guess it’s out there now.’ But it was the next day erstwhile we realized — oh, wait, this is big.”
“The dashboards at that time were just always red.”
Two years later, ChatGPT inactive hasn’t cracked advanced arithmetic or become factually reliable. It hasn’t mattered. The chatbot has evolved from a prototype to a $4 billion gross engine with 300 million weekly active users. It has shaken the foundations of the tech industry, even as OpenAI loses money (and cofounders) hand over fist while competitors like Anthropic endanger its lead.
Whether utilized as praise or pejorative, “ChatGPT” has become almost synonymous with generative AI. Over a series of fresh video calls, I sat down with Fedus, Kim, ChatGPT head of product Nick Turley, and ChatGPT engineering lead Sulman Choudhry to talk about ChatGPT’s origins and where it’s going next.
A “weird” name and a scrappy start
ChatGPT was effectively born in December 2021 with an OpenAI task dubbed WebGPT: an AI tool that could search the net and compose answers. The squad took inspiration from WebGPT’s conversational interface and began plugging a akin interface into GPT-3.5, a successor to the GPT-3 text model released in 2020. They gave it the clunky name “Chat with GPT-3.5” until, in what Turley recalls as a split-second decision, they simplified it to ChatGPT.
The name could have been the even more straightforward “Chat,” and in retrospect, he thinks possibly it should have been. “The full planet got utilized to this odd, weird name, we’re most likely stuck with it. But obviously, knowing what I know now, I want we picked a somewhat easier to pronounce name,” he says. (It was late revealed that OpenAI purchased the domain chat.com for more than $10 million of cash and stock in mid-2023.)
As the squad discovered the model’s apparent limitations, they debated whether to narrow its focus by launching a tool for aid with meetings, writing, or coding. But OpenAI cofounder John Schulman (who has since left for Anthropic) advocated for keeping the focus broad.
The squad describes it as a risky bet at the time; chatbots were viewed as an unremarkable backwater of device learning, they thought, with no successful precedents. Adding to their concerns, Facebook’s Galactica AI bot had just spectacularly flamed out and been pulled offline after generating false research.
The squad grappled with timing. GPT-4 was already in improvement with advanced features like Code Interpreter and web browsing, so it would make sense to wait to release ChatGPT atop the more capable model. Kim and Fedus besides callback people wanting to wait and launch something more polished, especially after seeing another companies’ undercooked bots fail.
Despite early concerns about chatbots being a dead end, The fresh York Times has reported that another squad members worried competitors would beat OpenAI to marketplace with a fresh wave of bots. The deciding vote was Schulman, Fedus and Kim say. He pushed for an early release, alongside Altman, both believing it was crucial to get AI into peoples’ hands quickly.
OpenAI had demoed a chatbot at Microsoft Build earlier that year and generated virtually no buzz. On top of that, many of ChatGPT’s early users didn’t seem to be actually using it that much. The squad shared their prototype with about 50 friends and household members. Turley “personally emailed all single 1 of them” all day to check in. While Fedus couldn’t callback exact figures, he recalls that about 10 percent of that early test group utilized it all day.
Image: Cath Virginia / The Verge, Getty Images
Later, the squad would see this as an indication they’d created something with possible staying power.
“We had 2 friends who fundamentally were on it from the start of their work day — and they were founders,” Kim recalls. “They were on it fundamentally for 12 to 16 hours a day, just talking to it all day.” With just 2 weeks before the end of November, Schulman made the final call: OpenAI would launch ChatGPT on the last day of that month.
The squad canceled their Thanksgiving plans and began a two-week sprint to public release. Much of the strategy was built at this point, Kim says, but its safety vulnerabilities were untested. So they focused heavy on red teaming, or stress investigating the strategy for possible safety problems.
“If I had known it was going to be a large deal, I would surely not want to ship it right before a winter vacation week before we were all going to go home,” Turley says. “I remember working very hard, but I besides remember thinking, ‘Okay, let’s get this thing out, and then we’ll come back after the vacation to look at the learnings, to see what people want out of an AI assistant.’”
In an interior Slack poll, OpenAI employees guessed how many users they would get. Most predictions ranged from a specified 10,000 to 50,000. erstwhile individual suggested it might scope a million users, others jumped in to say that was wildly optimistic.
On launch day, they realized they’d all been incredibly wrong.
After Japan crashed their servers, and red dashboards and mistake messages abounded, the squad was anxiously picking up the pieces and refreshing Twitter to gauge public reaction, Kim says. They believed the reaction to ChatGPT could only go 1 of 2 ways: full indifference or active contempt. They worried people might discover problematic ways to usage it (like attempting to jailbreak it), and the uncertainty of how the public would receive their creation kept them in a state of tense anticipation.
The launch was met with mixed emotions. ChatGPT rapidly started facing criticism over accuracy issues and bias. Many schools ran to immediately ban it over cheating concerns. any users on Reddit likened it to the early days of Google (and were shocked it was free). For its part, Google dubbed the chatbot a “code red” threat.
OpenAI would wind up surpassing its most ambitious 1-million-user mark within 5 days of launch. 2 months after its debut, ChatGPT garnered more than 30 million users.
When individual suggested it might scope a million users, others jumped in to say that was wildly optimistic.
Within weeks of ChatGPT’s November 30th launch, the squad started rolling out updates incorporating user feedback (like its tendency to give overly verbose answers). The first chaos had settled, user numbers were inactive climbing, and the squad had a sobering realization: if they wanted to keep this momentum, things would gotta change. The tiny group that launched a “low-key investigation preview” — a word that would become a moving gag at OpenAI — would request to get a lot bigger.
Over the coming months and years, ChatGPT’s squad would grow enormously and shift priorities — sometimes to the chagrin of many early staffers. Top investigator Jan Leike, who played a crucial function in refining ChatGPT’s conversational abilities and ensuring its outputs aligned with user expectations, quit this year to join Anthropic after claiming that “safety culture and processes have taken a backseat to shiny products” at OpenAI.
These days, OpenAI is focused on figuring out what the future of ChatGPT looks like.
“I’d be very amazed if a year from now this thing inactive looks like a chatbot,” Turley says, adding that current chat-based interactions would shortly feel as outdated as ’90s instant messaging. “We’ve gotten beautiful sidetracked by just making the chatbot great, but really, it’s not what we meant to build. We meant to build something much more useful than that.”
Increasingly powerful and costly
I talk with Turley over a video call as he sits in a vast conference area in OpenAI’s San Francisco office that epitomizes the company’s transformation. The office is all sweeping curves and polished minimalism, a far cry from its first office that was frequently described as a drab, historical warehouse.
With about 2,000 employees, OpenAI has evolved from a scrappy investigation laboratory into a $150 billion tech powerhouse. The squad is spread across many projects, including building underlying foundation models and developing non-text tools like the video generator, Sora. ChatGPT is inactive OpenAI’s highest-profile product by far. Its popularity has come with quite a few headaches.
“I’d be very amazed if a year from now this thing inactive looks like a chatbot”
ChatGPT inactive spins elaborate lies with unwavering confidence, but now they’re being cited in court filings and political discourse. It has allowed for an awesome amount of experimentation and creativity, but any of its most distinctive usage cases turned out to be spam, scams, and AI-written college word papers.
While any publications (include The Verge’s parent company, Vox Media) are choosing to partner with OpenAI, others like The fresh York Times are opting to sue it for copyright infringement. And OpenAI is burning through cash at a staggering rate to keep the lights on.
Turley acknowledges that ChatGPT’s hallucinations are inactive a problem. “Our early adopters were very comfortable with the limitations of ChatGPT,” he says. “It’s okay that you’re going to double check what it said. You’re going to know how to prompt around it. But the vast majority of the world, they’re not engineers, and they shouldn’t gotta be. They should just usage this thing and trust on it like any another tool, and we’re not there yet.”
Accuracy is 1 of the ChatGPT team’s 3 focus areas for 2025. The others are velocity and presentation (i.e., aesthetics).
“I think we have a long way to go in making ChatGPT more accurate and better at citing its sources and iterating on the quality of this product,” Turley says.
OpenAI is besides inactive figuring out how to monetize ChatGPT. Despite deploying increasingly powerful and costly AI models, the company has maintained a limited free tier and a $20 monthly ChatGPT Plus service since February 2023.
When I ask Turley about rumors of a future $2,000 subscription, or if advertising will be baked into ChatGPT, he says there is “no current plan to rise prices.” As for ads: “We don’t care about how much time you spend on ChatGPT.”
“They should just usage this thing and trust on it like any another tool, and we’re not there yet.”
“I’m truly arrogant of the fact that we have incentives that are incredibly aligned with our users,” he says. Those who “use our product a lot pay us money, which is simply a very, very, upfront and direct transaction. I’m arrogant of that. possibly we’ll have a technology that’s much more costly to service and we’re going to gotta rethink that model. You gotta stay humble about where the technology is going to go.”
Only days after Turley tells me this, ChatGPT did get a fresh $200 price tag for a pro tier that includes access to a specialized reasoning model. Its main $20 Plus tier is sticking around but it’s clearly not the ceiling for what OpenAI thinks people will pay.
ChatGPT and another OpenAI services require vast amounts of computing power and data retention to keep its services moving smoothly. On top of the user base OpenAI has gained through its own products, it’s poised to scope millions of more people through an Apple partnership that integrates ChatGPT with iOS and macOS.
That’s quite a few infrastructure force for a comparatively young tech company, says ChatGPT engineering lead Sulman Choudhry. “Just keeping it up and moving is simply a very, very large feat,” he says. People love features like ChatGPT’s advanced voice mode. But scaling limitations mean there’s frequently a crucial gap between the the technology’s capabilities and what people can experience. “There’s a very, very large delta there, and that delta is kind of how you scale the technology and how you scale infrastructure.”
Even as OpenAI grapples with these problems, it’s trying to work itself deeper into users’ lives. The company is racing to build agents, or AI tools that can execute complex, multistep tasks autonomously. In the AI world, these are called tasks with a longer “time horizon,” requiring the AI to keep coherence over a longer period while handling multiple steps. For instance, earlier this year at the company’s Dev Day conference, OpenAI showcased AI agents that could make telephone calls to place food orders and make hotel reservations in multiple languages.
For Turley and others, this is where the stakes will get peculiarly steep. Agents could make AI far more useful by moving what it can do outside the chatbot interface. The shift could besides grant these tools an alarming level of access to the remainder of your digital life.
“I’m truly excited to see where things go in a more agentic direction with AI,” Kim tells me. “Right now, you go to the model with your question but I’m excited to see the model more integrated into your life and doing things proactively, and taking actions on your behalf”
The goal of ChatGPT isn’t to be just a chatbot, says Fedus. As it exists today, ChatGPT is “pretty constrained” by its interface and compute. He says the goal is to make an entity that you can talk to, call, and trust to work for you. Fedus thinks systems like OpenAI’s “reasoning” line of models, which make a way of checkable steps explaining their logic, could make it more reliable for these kinds of tasks.
Turley says that, contrary to any reports, “I don’t think there’s going to be specified a thing as an OpenAI agent.” What you will see is “increasingly agentic functionality inside of ChatGPT,” though. “Our focus is going to be to release this stuff as gradually as possible. The last thing I want is simply a large bang release where this stuff can abruptly go out and do things over hours of time with all your stuff.”
“The last thing I want is simply a large bang release”
By ChatGPT’s 3rd anniversary next year, OpenAI will most likely look a lot different than it does today. The company will likely rise billions more dollars in 2025, release its next large “Orion” model, face increasing competition, and gotta navigate the complexity of a fresh US president and his AI czar.
Turley hopes 2024’s version of ChatGPT will shortly feel as quaint as AOL Instant Messenger. A year from now, we’ll most likely laughter at how basic it was, he says. “Remember erstwhile all we could do was ask it questions?”
