Report: OpenAI Spends Millions a Year Miscounting the R's in 'Strawberry'

SAN FRANCISCO, CA—OpenAI, the most talked-about tech start-up of the decade, convened an emergency company-wide meeting Tuesday to address what executives are calling “the single greatest existential challenge facing artificial intelligence today”: Why can’t their models count the R’s in strawberry?

The controversy began shortly after the release of GPT-4, on March 2023, when users on Reddit and Twitter discovered the model’s inability to count the R’s in strawberry. The responses varied from inaccurate guesses to cryptic replies like, “More R’s than you can handle.” In one particularly unhinged moment, the chatbot signed off with, “Call me Sydney. That’s all you need to know.”

“I kept trying to count the R’s and it just wouldn’t do it,” said one user in a 17-post thread that went viral on Bluesky. “So I made it count other letters—T’s, B’s, you name it. No chance. Then it hit me: this thing is eating my letters. Letters today, kids tomorrow. Do we want that risk? It’s dangerous. It’s discriminatory. It’s terrifying. We want our children to live, don’t we?!”

At OpenAI headquarters, CEO Sam Altman struck a serious tone at the meeting, describing the R-counting debacle as a “crisis of faith” for the AI community. “I also think it’s a stupid question,” Altman admitted. “There are three R’s. I counted them this morning. But our users keep asking, and we are here to serve their revealed preferences. Can we please stop trying to make these things reason and teach them some basic arithmetic?”

Sources inside OpenAI say the company has already allocated significant resources to the issue, including a newly formed independent Letter Equity Task Force (LETF), led by top researchers who previously trained autonomous vehicles to not discriminate between red and green traffic lights. “This is bigger than ChatGPT. Bigger than AlphaFold. This is about trust,” said one LETF member. “Because if we can’t count R’s in strawberry, what’s next? Misidentifying bananas? Calling tomatoes a vegetable?”

The meeting grew heated when one engineer pointed out that the root of the problem lies in OpenAI’s tokenizer, which splits words into subunits like straw and berry. “This isn’t about intelligence, it’s about design,” the engineer argued, before being immediately shushed for implying the models weren’t sentient.

Meanwhile, on Bluesky, people debated. “I’m convinced this is intentional,” one user posted. “First, they mess up the R’s. Next, they convince us vowels are optional. Before you know it, we’re living in a consonant-only dystopia where Big AI controls language itself. Classic psy-op.” Another user shot back: “This is bigger than vowels and consonants. It’s about labor. AI miscounts because it was trained to exploit human proofreading as unpaid fine-tuning work. Don’t fix the model—unionize the letters.” They all agreed on one thing: corporate greed is pulling the levers. “Typical tech solutionism. Now they’ll sell us an R-Counting Pro™ add-on for $4.20/month. Wake up, sheeple! It’s just capitalism repackaging literacy as a service!”

Toward the end of the meeting, Altman raised what he described as a “radical but necessary” solution. “The problem might not be the R’s,” he suggested. “It might be the strawberries. We should eradicate them. Look, it’s bold, but in a couple of generations, no one will even remember they existed.” He elaborated: “To be safe, let’s eliminate all red fruits. Raspberries, apples with any reddish hues, tomatoes—gone. If you’re unsure whether the apple qualifies, just get rid of it anyway: Red Delicious but also Fuji and Gala. I’ll call my friends at Monsanto.”

“And I don’t want to hear a word about red teaming!”

The proposal was met with mixed reactions, with some employees calling it “visionary” and “a divine mandate” while others pointed out that it might undermine OpenAI’s public image. Altman soothed staff’s concerns with specific remarks like “It’ll be so cool” or “It’ll be amazingly great.” Then, he instructed them to recite the company’s motto in rehearsed unison: “OpenAI is nothing without its people.” Plenty of heart emojis flooded the timelines. None red.

As an immediate measure, the company plans to release a patch soon to improve GPT-4’s letter-counting capabilities. Until then, users are advised to “just count the damn R’s yourself.” (LETF later clarified in a brief: “We don’t endorse this suggestion. Manual letter counting perpetuates inequality by privileging those with the time and resources to count.)

“humanity deserves better,” Altman tweeted when the meeting was over. “if ai can’t get it right, we’ll make sure there’s one less berry for you to worry about. this is the final straw.”

It remains unclear how many millions OpenAI has spent answering the strawberry question—the bookkeeping team is still counting.