AI Detectors
One thing I have pondered for some time is how artificial intelligence (AI) will shape the future of online content. Many people initially showcased ChatGPT’s generative AI on YouTube and narrated some scene written by AI and suggest these are not so different from a person. This is because in 2024 we have two forms of media that is at first glance difficult to distinguish between being generated from AI or man-made: writing and images.
There are two ways of detecting AI generated content that are popular, these are: visible or invisible watermarking, and AI detection tools. Detection tools are simply software for deciphering digital content and calculating if any piece of work is original or AI generated. AI detectors determine this through complexity and similarity to other works. An evaluation is usually given in percentage and highlights the affected areas visually.
Undoubtedly cheating is on the rise in schools because of generative AI, but the job market is being saturated with similar problems too, as individuals have been using a mass-apply tool for jobs that involves the use of automated AI software. It is no wonder job applications are becoming less human friendly as a result.
"Auto_Jobs_Applier_AIHawk" 18k stars on GitHub:
English learners #
One thing about AI is how it writes so terribly to the trained eye. There are many repeated catchphrases and Americanisms used by the default configuration of the ChatGPT programme that it becomes more noticeable over repeated exposure. This is useful in the UK where you can decipher how sincere your local business is about their reply by use of non-continental language.
I fear using too much AI will impact English learners the same way US television ruined my vocabulary. The reverse is also true: if you are a savant writer or unlucky to be a false positive then AI detectors will suggest your writing is possibly unoriginal and as a result you could lose your reputation. In some cases you miss out on a grade and may struggle to prove what you wrote was in fact original thought, this was the case for Moira Olmsted in University in 2023.
What are a few problems we facing because of generative AI? #
- Search ranking algorithms will change frequently and cause businesses a few problems.
- Popular communication applications will soon have inline AI writing tools. This may be socially depriving.
- “Prompt hacking” is still a weakness of AI and may leak personal data.
What I thought would happen through the introduction of generative AI #
- Smart devices would be more intelligent.
- More creative problems and jobs given to people that robots cannot yet do.
- The use of AI to rule out natural human bias when making tough decisions and making managers, administrators, and executives redundant.