Chat GPT wants everyone to know youโ€™re using Chat GPTโ€ฆ ๐™€๐™“๐˜พ๐™€๐™‹๐™ ๐™”๐™Š๐™! Most people are using AI in some form or another for word processing. It could be for inspiration for meal prep or youโ€™re using it to write full legal documents and contracts. Either way itโ€™s part of the zeitgeist and itโ€™s not going anywhereโ€ฆ Especially on social media. More and more as a business community we are being told we need to do more on socials but if you are time poor it can seem like a drain on resources. Chat GPT has offered an almost perfect solution to this by providing instant content without much time or effort. This is clear when you scroll through your news feed and see the constant use of the words โ€œMeaningful, fostering, no-fluff, whether, etc, etc, etcโ€ For us who use socials everyday to promote our businesses, itโ€™s become a daily game of spotting these words or the obnoxiously long hyphen which triggers the Ick response and a swift skim of the thumb to instantly dismiss the poorly constructed post. But Open AI knows this is obvious so for the more eagle eyed users of Chat GPT, they have built in new techniques to keep signalling that your text has been written by AI. (๐˜ž๐˜ฉ๐˜ช๐˜ญ๐˜ด๐˜ต ๐˜ธ๐˜ณ๐˜ช๐˜ต๐˜ช๐˜ฏ๐˜จ ๐˜ต๐˜ฉ๐˜ช๐˜ด ๐˜ ๐˜ต๐˜ฉ๐˜ฐ๐˜ถ๐˜จ๐˜ฉ๐˜ต ๐˜ข๐˜ฃ๐˜ฐ๐˜ถ๐˜ต ๐˜ฑ๐˜ถ๐˜ต๐˜ต๐˜ช๐˜ฏ๐˜จ ๐˜ข ๐˜ด๐˜ถ๐˜ฃ-๐˜ฉ๐˜ฆ๐˜ข๐˜ฅ๐˜ช๐˜ฏ๐˜จ ๐˜ฐ๐˜ง โ€˜๐˜ธ๐˜ฉ๐˜ข๐˜ต ๐˜บ๐˜ฐ๐˜ถโ€™๐˜ญ๐˜ญ ๐˜ญ๐˜ฆ๐˜ข๐˜ณ๐˜ฏโ€™ ๐˜ฐ๐˜ณ ๐˜ธ๐˜ฉ๐˜ข๐˜ต ๐˜ฅ๐˜ฐ๐˜ฆ๐˜ด ๐˜ต๐˜ฉ๐˜ช๐˜ด ๐˜ฎ๐˜ฆ๐˜ข๐˜ฏ ๐˜ข๐˜ด ๐˜ข ๐˜ซ๐˜ฐ๐˜ฌ๐˜ฆ ๐˜ฃ๐˜ถ๐˜ต ๐˜ฅ๐˜ช๐˜ฅ๐˜ฏโ€™๐˜ต ๐˜ต๐˜ฉ๐˜ช๐˜ฏ๐˜ฌ ๐˜ช๐˜ต ๐˜ธ๐˜ฐ๐˜ถ๐˜ญ๐˜ฅ ๐˜ญ๐˜ข๐˜ฏ๐˜ฅ.) ๐˜ข๐˜ฏ๐˜บ๐˜ฉ๐˜ฐ๐˜ฐโ€ฆ So the creators of Chat GPT have sneakily built in things called โ€˜Watermarksโ€™ to keep signalling they have written text even after youโ€™ve removed the tell tale signs. Watermarks are basically like hitting your space bar on your keyboard and not being able to see the space. They are usually undetectable to the naked eye but form a gap in the text that you wouldnโ€™t be aware of. Social media algorithms are trained to recognise these watermarks and instantly recognise if itโ€™s not authentic content. This could have a major effect on the reach of your social media posts. Try it for yourself. Create a piece of text in Chat GPT and copy & paste it into a watermark detector and see how many you find. See the pictures attached as an example. Comment below with the number of watermarks you find...

Posted by Chris at 2025-09-16 06:30:04 UTC