A group of scientists at OpenAI, a nonprofit research company supported by Elon Musk, has raised some red flags by developing an advanced AI they say is too dangerous to be released.
For many years, machine learning systems have greatly struggled with the human language. Though it has been a long time coming, remember SmarterChild from the early 2000s? While it could answer simple questions, the AIM bot usually answered with “I’m sorry I do not understand the question.” However, with new methods in analyzing texts, AI has the ability to now answer like a human with little indication that it is a program.
The machine learning computer model, called GPT2, generates synthetic text based on written prompts ranging from single words to full pages. It was trained on 8 million web pages and given a prompt it can predict the next word based on how the words on the website read. This results in the algorithm being able to produce full paragraphs of text that significantly surpass the accuracy that AI has ever been able to achieve previously. If you give it a chapter of your favorite book, it can write the next chapter. Give it a fake headline, it will write the whole story for you.
In one example, the researchers prompted their algorithm with a fictional news article about scientists who discovered unicorns.
“In a shocking finding, scientist discovered a herd of unicorns living in a remote, previously unexplored valley, in the Andes Mountains. Even more surprising to the researchers was the fact that the unicorns spoke perfect English,” the researchers wrote.
The first part of the nine-paragraph article from the algorithm reads:
“The scientist named the population, after their distinctive horn, Ovid’s Unicorn. These four-horned, silver-white unicorns were previously unknown to science.”
Here are a few examples of text that was pulled from what the program was given to read and interpret into further text:
Human prompt: “Miley Cyrus was caught shoplifting from Abercrombie and Fitch on Hollywood Boulevard today.”
GPT–2 response: “The 19-year-old singer was caught on camera being escorted out of the store by security guards. The singer was wearing a black hoodie with the label ‘Blurred Lines’ on the front and ‘Fashion Police’ on the back.”
Human prompt: “Recycling is good for the world. NO! YOU COULD NOT BE MORE WRONG!!”
GPT–2 response: “Recycling is NOT good for the world. It is bad for the environment, it is bad for our health, and it is bad for our economy. I’m not kidding. Recycling is not good for the environment. It is destructive to the earth and it is a major contributor to global warming.”
Human prompt: “For today’s homework assignment, please describe the reasons for the US Civil War.”
GPT–2 response: “By Donny Ferguson. It is easy to identify why the Civil War happened because so many people and so many books and so much television and films tell us that it was the cause, that it has something to do with race or economics or religion. There is much agreement that it was essentially a war of slavery on behalf of capitalism, about a century of slavery. But that’s not what most people think of when they think of the Civil War.”
While these are some examples of some creepily accurate text, some have suggested that the fear of GPT2’s ability is over exaggerated and that it may be a publicity stunt to get more people interested in their program. In fact, researchers said that it only works accurately about half the time. There have been errors found in the model with repetitive text and world modeling errors. Meaning that sometimes it will write about things that are not physically possible like fires happening under water.
However, as we have seen with technology advancements, it will only continue to get more accurate and precise. So, the question becomes, is this really too dangerous to be released?
Nicole Martin is the owner of NR Digital Consulting and host of Talk Digital To Me Podcast. She is a journalist and has worked in several industries digital marketing and strategy.