OpenAI, the non-profit organization founded by Elon Musk, has created Chat GPTan artificial intelligence which thanks to its advanced functions already has more than a million users worldwide and which could make search engines obsolete, as well as numerous jobs.
The concerns are well founded and were also confirmed by a Gmail developer, who tweeted last week that “in a year or two Google may become useless. AI will take down the search engine results page, which generates the most revenue”.
ChatGPT has attracted the attention (and concern) of many for its advanced features, ranging from the instantaneous composition of complex essays to writing source code, passing through the drafting of marketing proposals or interior design projects. The artificial intelligence is based on Reinforcement Learning from Human Feedback, an algorithm that relies on human responses to “create a new model that is presented in an intuitive UI, with a certain amount of memory”. This aspect makes ChatGPT much more humane than search engines, despite having at its disposal a mass of data equal to a supercomputer.
In a example cited by the New York Post we are talking about a trivial question asked by users to Google: “what is the maximum dosage of Vitamin D per day?”. Big G returns as a message the guidelines that I refer to HealthLine, while the AI has formulated a more thorough and broader answer.
However, a Twitter user also asked ChatGPT to write “a haiku from the point of view of a copywriter who feels sad that artificial intelligence could diminish the value of the written word”. The AI replied like this: “words on a screen, now just a blur, the machine takes the pen”.
The NYPost itself does not dwell on creative uses, which in any case could potentially redefine the world economy, but also malicious ones: for example is capable of program malware and phishing emails.
