ChatGPT trained to lie

Words: Laura Jones Image: Getty Images

“They’re training AI to lie … to comment on some things and not others … to withhold …” Elon Musk spoke avidly about Chat GPT and its failure to be accurate and honest to Tucker Carlson at The Media Research Centre. He criticised Microsoft for training the system to lie – ChatGPT is known for its flawed and untrue versions of reality as is Microsoft’s Bard. ChatGPT is a so-called large language model, which is trained on an enormous amount of data to interpret text and respond to prompts in a human-like manner. Both confidently flout fake facts and as such, aren’t valid research tools that can be relied on. Both make the declaration of the fact upfront before users even begin. It’s point of use is differently abled. Musk however an ace at understanding how digital media, particularly social media aggregates commentary to alter and influence perceptions. He excels at it for his own endeavours. Musk is the serial entrepreneur, and investor founding PayPal, Tesla, and SpaceX. Having seriously disrupted the banking sector, the car industry and space travel, he is artful in his craft.

Specially since he has plans to introduce his own AI rival to ChatGPT called TruthGPT. His next step is to reportedly leverage his new AI company off Twitter to accomplish this and ultimately wear the crown of innovation. After dropping out of AI a few years ago, albeit being one of the founders, Musk believes that AI is currently dangerous. 

DANGERS

Speaking in an interview with Tucker Carlson Musk says, “AI is more dangerous than, say, mismanaged aircraft design or production maintenance or bad car production, in the sense that it is, it has the potential — however small one may regard that probability, but it is non-trivial — it has the potential of civilization destruction”. In the wrong hands, or even in the right hands, all hell could break loose. We’re not talking only about the issue an 11-year-old using the platform to write his projects handing them in as his own (Yes, the platform will write to the level of an 11-year old if you prompt it).

The New York Times writes that tech leaders have criticized an “out-of-control race” to develop more advanced artificial intelligence. In addition, more than 1 000 technology leaders and researchers, including Musk, have urged artificial intelligence labs to pause development of the most advanced systems, warning in an open letter that A.I. tools present “profound risks to society and humanity.”

REGULATION

Sky News reports that the Chinese government has published draft rules outlining how generative AI services would likely have to adhere to the same tight restrictions as the rest of the internet in China. In March, Italy became the first country to outright ban ChatGPT while the country’s data protection authorities investigated its collection of user information. 

In the EU law enforcement agency Europol also warned ChatGPT may be used by criminals and to spread disinformation. And the White House is also inviting public feedback on how AI should be regulated in the US, having unveiled its “blueprint for an AI bill of rights” late last year to protect jobs and privacy.

Reports confirm that Musk has already incorporated a new business called X.AI Corp, according to a US business filing, listing Musk as its director and his adviser, Jared Birchall, as secretary. This comes after Twitter has merged into a firm named XCorp.

Tweeting on AI Musk adds, “ Even benign dependency on AI/Automation is dangerous to civilization if taken so far that we eventually forget how the machines work.”