AI-content generation has a long way to go.
What you need to know
- Microsoft published an AI-generated article highlighting key tourist attractions in Ottawa, Canada on Microsoft Start.
- The article has since been pulled down after recommending Ottawa Food Bank among the top three places to visit.
- The company is currently investigating the matter to determine how the piece made it through the review process and being published.
While generative AI has gained massive popularity this year, people still have reservations towards the tool. Privacy and security concerns top the list, but the issue of accuracy still lurks among many users.
And as it seems, it remains a major issue. Microsoft recently posted an article on Microsoft Start (MSN) dubbed "Headed to Ottawa? Here's what you shouldn't miss!" And while you'd expect this list to be packed with cool and fun places you can visit while in Ottawa, Canada, the third option on the list was off-putting.
This is because it recommended the Ottawa Food Bank as one of the top tourist attractions with the following caption: "People who come to us have jobs and families to support, as well as expenses to pay. Life is already difficult enough. Consider going into it on an empty stomach," as spotted by Paris Marx on X (formerly Twitter).
Alongside, the bizarre recommendation was The Winterlude Festival, National War Memorial, as well as the consideration of going to an Ottawa Senators hockey game. However, when you click on the link to the original article on Microsoft Start, you'll be met by a message indicating that the page no longer exists. A clear indication that Microsoft pulled down the article. The Verge has since uploaded the full article on Imgur, so you can also check it out from there.
“This article has been removed and we are investigating how it made it through our review process," said Jeff Jones, a senior director at Microsoft while speaking to The Verge on the issue.
In recent studies and investigations carried out by researchers, AI-powered chatbots are seemingly getting worse. For example, ChatGPT is reportedly getting dumber, despite numerous updates and improvements. OpenAI, which makes ChatGPT, could potentially be on the verge of bankruptcy as well.
You might recall that Microsoft laid off a bunch of its MSN journalists in 2020 while transitioning to AI-generated content, in the face of the COVID-19 pandemic.
Samantha Koziara, Ottawa Food Bank's communication manager shared the following sentiments with The Verge pertaining to the article:
The ‘empty stomach’ line is clearly insensitive and didn’t pass by a (human) editor. To my knowledge, we haven’t seen something like this before — but as AI gets more and more popular, I don’t doubt an increased number inaccurate / inappropriate references will be made in listicles such as this. This simply highlights the importance of researchers, writers, and editors… of the human variety.
AI needs more tuning
Evidently, there's still a lot of work to be done on the AI front if organizations and companies want to cut down on staff and replace them with Artificial Intelligence.
The technology is still far from perfection, with instances of chatbots giving wrong information or outrightly being rude to users. Additionally, the FTC launched an investigation looking into ChatGPT, to determine whether OpenAI had broken any consumer protection laws.
While the FTC can't make rules that govern how AI tools are used, it could still make recommendations that would have a great impact. As such, Microsoft is also likely to be on the receiving end given its recent interest and investment in AI.
- aum
- 1
Recommended Comments
There are no comments to display.
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.