Risky Business: What An Overreliance On AI Can Mean For Companies As ChatGPT Goes Down

Recent months have seen a rising tide of public voices warning against the risks of fast-growing AI technology.

On Thursday, a global outage of ChatGPT served as a warning of what an overreliance on artificial intelligence could mean for business operations.

From the short-term danger of losing millions of jobs, all the way to becoming an existential risk to humanity, AI has quickly climbed to the top amongst concerns from politicians and scientists, with one expert calling the issue "more urgent" than climate change.

Meanwhile, investors are showing their support for AI technology.

On Thursday, stocks from NVIDIA Corporation NVDA jumped 26% after the company's latest earnings call, which helped cement its place as the "unquestionable AI leader."

The Risks Of Too Much AI In The Workplace

According to a paper released last week by OpenAI's research arm, "80% of the U.S. workforce could have at least 10% of their work tasks affected by the introduction of GPTs," while around a fifth of workers may see at least half of their tasks impacted.

These impacts are expected across the wage ladder and across industries, "with higher-income jobs potentially facing greater exposure."

Read Also: 61% Of Americans Think AI Could Spell The End Of Humanity, Echoing Elon Musk’s Warning: Report

While even the White House has recently jumped into the AI-prepping trend, one concern has eluded most disaster predictions: the potential risks of an overreliance of AI in the business world.

Thursday's ChatGPT outage lasted a little over one hour from 7:23 A.M. to 8:27 A.M Eastern Time, according to Gizmodo.

"Elevated database CPU usage was impacting the site over the last several hours, and the underlying issue has been addressed," said OpenAI, promising to "ensure the issue does not reoccur." 

ChatGPT is normally accessed through OpenAI's server. While there are ways to run the program locally on a private server, OpenAI's business model is built around subscriptions, and most businesses —even those with enough infrastructure— will benefit from not having to host the AI in their own servers.

This means that a ChatGPT outage could eventually mean an operations outage for most businesses, as companies continue to rely more and more on this type of software.

Businesses are already using OpenAI's API to build chatbots and digital assistants as well as a plethora of other tasks like data classification, text analysis and coding assistance.

Yet many more businesses are beginning to use ChatGPT manually to complement their operations.

The app broke the record for achieving the fastest-growing user base in history, reaching the 100 million mark in just over two months. It took TikTok nine months and Instagram — owned by Meta Platforms Inc META — two-and-a-half years to get there.

Companies are beginning to rely on ChatGPT's text-generation abilities to brainstorm ideas, write copy, create social media content, write presentations, emails and more.

While these tasks are still being run by humans that could replace the AI in case of failure, once the technology becomes a part of business as usual, a ChatGPT outage could mean a drop in productivity across the board.

An overreliance on automated text generation could also lead companies into legal trouble. 

From so-called "hallucinations," where ChatGPT will output false information, to gender or race-based discrimination, delegating too much responsibility into ChatGPT-like software could lead to companies to commit unwanted errors that can turn into a litigious hazard.