Jay Smith
United Kingdom
๐—œ๐˜€ ๐—”๐—œ ๐—ฆ๐—น๐—ผ๐—ฝ ๐—ž๐—ถ๐—น๐—น๐—ถ๐—ป๐—ด ๐—ง๐—ต๐—ฒ ๐—œ๐—ป๐˜๐—ฒ๐—ฟ๐—ป๐—ฒ๐˜? //๐‘๐‘Ž๐‘Ÿ๐‘ก ๐Ÿท ๐‘œ๐‘“ ๐Ÿน - ๐‘ ๐‘’๐‘’ ๐‘๐‘œ๐‘š๐‘š๐‘’๐‘›๐‘ก๐‘  ๐‘“๐‘œ๐‘Ÿ ๐‘๐‘Ž๐‘Ÿ๐‘ก๐‘  ๐Ÿธ & ๐Ÿน AI Slop - Weโ€™ve all seen it in one form or another; low quality marketing imagery, poorly written social media comments and weird Instagram videos that defy logic. In todayโ€™s post, Iโ€™ll look at how the tidal wave of AI slop is impacting the internet, and how that translates into my investment approach. ๐™’๐™๐™–๐™ฉ ๐™ž๐™จ ๐˜ผ๐™„ ๐™Ž๐™ก๐™ค๐™ฅ? Broadly speaking, AI slop is low quality content which has been produced by AI. The format can vary: โ€ข Viral videos of fake disasters โ€ข Baffling Social Media Comments โ€ข Buzzword filled marketing emails โ€ข Strange pictures that donโ€™t quite work Since so many AI products are free, and so much of our online world has little filtration or limits for content being uploaded, AI slop is now everywhere. The big problem is that this new wave of AI content is often misleading, irrational, biased, or wrong. ๐™๐™๐™š ๐˜ฟ๐™š๐™–๐™ฉ๐™ ๐™ค๐™› ๐˜พ๐™ค๐™ฃ๐™ฉ๐™š๐™ฃ๐™ฉ ๐˜พ๐™ง๐™š๐™–๐™ฉ๐™ž๐™ค๐™ฃ ChatGPT 3.5 was launched in November 2022, but it only took around 2 years for waves of AI generated content to start hitting the internet. AI slop has infiltrated every medium of content online. ๐‘ฑ๐’๐’–๐’“๐’๐’‚๐’๐’Š๐’”๐’Ž While most major news outlets are actively avoiding using AI too much because of reputational risk, there has been a significant rise in โ€œPink Slimeโ€ journalism - organizations that scrape the internet for stories, then re-phrase and regurgitate them without any fact checking. ๐‘ด๐’–๐’”๐’Š๐’„ AI-Generated bands recently caught headlines and some backlash on $SPOT (Spotify Technologies SA) , amid speculation that the company was competing with real artists by generating music to fill out their playlists. While much of it can be considered Slop, some of the AI bands have grown a significant following on the platform. ๐‘จ๐’“๐’• Artists have complained that their style has been stolen by AI and is being used to generate more personalizable artwork. In April the trend of โ€œGhiblifyingโ€ profile pictures - generating them in the style of Studio Ghibli art swept social media. Ironically, Studio Ghibli founder Hayao Miyazaki is well known for having strong views against AI assisted art. ๐‘ฝ๐’Š๐’…๐’†๐’ ๐‘ฎ๐’‚๐’Ž๐’†๐’” As an avid gamer myself, Iโ€™ve witnessed the influx of AI Slop into gaming. Earlier this year it was reported that the number of steam titles disclosing generative AI usage has grown 800% since last year. Meanwhile one of the biggest hit titles of the year - Peak, is facing a different problem. Their game is being copied and cloned using AI and is flooding online stores. Valve has been battling the surge of copycat games being uploaded on a daily basis. ๐‘ฉ๐’๐’๐’Œ๐’” Itโ€™s never been more difficult for authors to be discovered. Recommendation algorithms on Amazon are struggling to sort through it all, and customers are finding un-edited ChatGPT prompts in books theyโ€™ve purchased. Whatโ€™s worse is that AI is being used to read the books, then consolidate them by โ€œsummary farmsโ€, which sell shortened versions. ๐™„๐™ข๐™ฅ๐™–๐™˜๐™ฉ ๐™ค๐™ฃ ๐™Š๐™ช๐™ง ๐™Š๐™ฃ๐™ก๐™ž๐™ฃ๐™š ๐™Ž๐™ฅ๐™–๐™˜๐™š๐™จ Fake profiles are now flooding our online spaces, from social media trolls, and content creators who donโ€™t exist, to misleading reviews and generically named customer service bots in place of humans when you try to contact support for a product or service. These online AI personas can be used to influence people politically, create โ€˜engagement baitโ€™ to generate income, or be deployed into comments sections and reviews to overwhelm recommendations or lure in new users or customers. Thereโ€™s also a potential problem brewing for the AIโ€™s we use on a daily basis as a replacement for search engines. As more and more content online is being generated by AI, that same content is being consumed by AI again for training. This risks creating a dangerous feedback loop and echo chambers that reinforce inaccurate content. //๐‘’๐‘›๐‘‘ ๐‘œ๐‘“ ๐‘๐‘Ž๐‘Ÿ๐‘ก ๐Ÿท ๐‘œ๐‘“ ๐Ÿน - ๐‘ ๐‘’๐‘’ ๐‘๐‘œ๐‘š๐‘š๐‘’๐‘›๐‘ก๐‘  ๐‘“๐‘œ๐‘Ÿ ๐‘๐‘Ž๐‘Ÿ๐‘ก๐‘  ๐Ÿธ & ๐Ÿน $GOOG (Alphabet) $MSFT (Microsoft) $META (Meta Platforms Inc) $RDDT (Reddit Inc)
20 of 32
2 replies
2 replies
2 replies
1 reply
3 replies
1 reply
null
.