Page 1 of 1

DALL-E 2 art board. Source: Pinterest

Posted: Thu Feb 06, 2025 3:32 am
by Rina7RS
Official reason?

According to the OpenAI blog announcement, the team is concerned that GPT-2 could be used to write scam emails or generate fake news. This makes perfect sense. After all, with great power comes great responsibility thanks, Peter Parker.

In 2019, OpenAI made another controversial decision to transform into a “capped profit” organization and formed OpenAI LP, a “hybrid of for-profit and nonprofit.”

DALL·E, GPT-3, and ChatGPT 2021–
In 2021, OpenAI released DALL-E, an AI that uses a similar architecture spain mobile database to GPT-2. DALL-E — a portmanteau of WALL-E and Spanish surrealist artist Salvador Dalí — can generate realistic images, but instead of generating text, it seems to come out of thin air.


In 2022, OpenAI pushed the envelope again with GPT-3. An iteration of the first two models, GPT-3 was fed 45TB of text data, which was converted into 175B of parameters. It is smarter, faster, and scarier than anything we’ve seen before.

To achieve this goal, Microsoft designed a supercomputer for OpenAI that includes 285,000 CPU cores and 10,000 GPUs. It is also ranked 5th in the Top500 supercomputer ranking.

The success of GPT-3 has spawned another monster in disguise.

In November 2022, OpenAI released ChatGPT, a language model chatbot built on GPT-3.

One of the most amazing aspects of ChatGPT is its ability to understand context — the chatbot can generate answers and adapt based on the conversation history. This means you can “train” ChatGPT in a conversation thread to get more accurate answers.