{"id":96,"date":"2024-06-02T18:06:32","date_gmt":"2024-06-02T18:06:32","guid":{"rendered":"https:\/\/artificial-intelligence.news\/?p=96"},"modified":"2024-06-02T18:06:32","modified_gmt":"2024-06-02T18:06:32","slug":"is-ai-over-hyped","status":"publish","type":"post","link":"https:\/\/artificial-intelligence.news\/?p=96","title":{"rendered":"Is AI over hyped?"},"content":{"rendered":"\n

AI has been at the forefront of technological advancements in the past year, and with good reason. Generative text, video and music have for the first time lived up to the promises made decades ago. The results have been mind blowing, but their limitations are often overlooked and sometimes used to make overblown claims.<\/p>\n\n\n\n

<\/p>\n\n\n\n

Google Gemini Gimmick<\/h2>\n\n\n\n

We have already mentioned how the demo for Google Gemini<\/a>, where a rubber duck was described as it was being drawn, was faked. The responses from the AI were pre-recorded where it was presented still images of the drawing. This is contrary to the demo video, which makes it appear that Gemini is recognising what is happening in real time.<\/p>\n\n\n\n

<\/p>\n\n\n\n

Devin Debunked<\/h2>\n\n\n\n

More recently we covered Devin – The AI developer<\/a>, which was shown completing an Upwork freelancing project. But many of it’s claims were debunked by the Internet of Bugs<\/a> YouTuber. The video shows the task was to provide instructions for running a model on AWS. Instead Devin misses this main objective, and instead runs the model locally and provides a report:<\/p>\n\n\n\n

\"\"<\/figure>\n\n\n\n

<\/p>\n\n\n\n

Another claim of the Devin demo was it fixing a bug in the repository provided by the Upwork client. However, when searching the repository for the filename in the video, no results show up. This shows that the “bug” was created by Devin itself, contrary to what is suggested in the demo:<\/p>\n\n\n\n

\"\"<\/figure>\n\n\n\n

<\/p>\n\n\n\n

The quality of the code is also low, with it missing instructions for running the project in the Readme file and instead building it’s own botched together version. As well as using poor code constructs to read data into a buffer. And the use of nonsensical commands, such as this one where it reads the first five lines of a file and then uses those to read the last five lines:<\/p>\n\n\n\n

\"\"<\/figure>\n\n\n\n

<\/p>\n\n\n\n

Amazon AI checkout (powered by humans)<\/h2>\n\n\n\n

Another notable example is Amazon’s claim of using AI to create checkout-less shops, was in fact backed by a team of 1,000 workers in India manually reviewing a reported 70%<\/a> of all checkouts. As a result Amazon is ditching these “Just Walk Out” stores<\/a>, in favour of a less high tech solution with scanners embedded into shopping carts.

This has some scary paralles to the Mechanical Turk<\/a> scam from the 18th century, where a human hiding inside a chess machine was claimed to be fully automated. Funnily enough, Amazon Mechanical Turk<\/a> is an actual service that allows you to pay humans cents to complete thousands of routine tasks.<\/p>\n\n\n\n

\"\"<\/figure>\n\n\n\n

Conclusion<\/h2>\n\n\n\n

These three cautionary tales are examples of not believing everything you hear, even when the sources are reputable companies. It emphasises the importance of only believing what you can try with your own hands, rather than relying on recorded demos or published reports.<\/p>\n\n\n\n

<\/p>\n","protected":false},"excerpt":{"rendered":"

AI has been at the forefront of technological advancements in the past year, and with good reason. Generative text, video and music have for the first time lived up to the promises made decades ago. The results have been mind blowing, but their limitations are often overlooked and sometimes used to make overblown claims. Google […]<\/p>\n","protected":false},"author":2,"featured_media":101,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"site-container-style":"default","site-container-layout":"default","site-sidebar-layout":"default","disable-article-header":"default","disable-site-header":"default","disable-site-footer":"default","disable-content-area-spacing":"default","footnotes":""},"categories":[1],"tags":[],"class_list":["post-96","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-blog"],"_links":{"self":[{"href":"https:\/\/artificial-intelligence.news\/index.php?rest_route=\/wp\/v2\/posts\/96","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/artificial-intelligence.news\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/artificial-intelligence.news\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/artificial-intelligence.news\/index.php?rest_route=\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/artificial-intelligence.news\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=96"}],"version-history":[{"count":1,"href":"https:\/\/artificial-intelligence.news\/index.php?rest_route=\/wp\/v2\/posts\/96\/revisions"}],"predecessor-version":[{"id":102,"href":"https:\/\/artificial-intelligence.news\/index.php?rest_route=\/wp\/v2\/posts\/96\/revisions\/102"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/artificial-intelligence.news\/index.php?rest_route=\/wp\/v2\/media\/101"}],"wp:attachment":[{"href":"https:\/\/artificial-intelligence.news\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=96"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/artificial-intelligence.news\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=96"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/artificial-intelligence.news\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=96"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}