How GPT-3 Works:
After inputting text, the model predicts what the input following it should be well because its "memory" contains all text ever published by humans on the internet. As GPT-3 was trained on a dataset of half a trillion words – an archive of the internet called the Common Crawl – the model can also identify linguistic patterns within the text and more, allowing it to understand more nuanced human text.
Naturally, the internet has more or less lost its mind over the new tech, which some claim also shows some self-awareness.
Vercel CEO Guillermo Rauch says it represents “general intelligence” "and "a building block for a new category of apps," adding that it feels like “Google as an API.”
While many are excited by the possibilities it creates, some developers are nervously wondering if this will render their jobs obsolete. As Analytics India Magazine reports, software development is one of the career paths GPT-3 most threatens. Indian entrepreneur Vijay Anands, the founder of Proto.in - India's biggest startup showcase platform – tweeted the tool represents the "Next level of #nocode" – a movement that already is worrying developers and threatens to change the field.
While it certainly does represent "the next level" of the no-code movement, are developer jobs now really as in jeopardy as some now think because of GPT-3?
The answer: It's complicated.