yann lecun head of ai for meta

The AI-powered chatbot created by OpenAI – ChatGPT – has grabbed the headlines for weeks as the public has been amazed by how capable the software is to come up with seemingly clever answers and even successfully passing some complicated tests.

However, for the Chief AI scientist of Meta Platforms (META), Yann LeCun, ChatGPT can hardly be considered innovative as many other companies have already created similar technologies, although not once has delivered the seamless user experience that this chatbot has.

In a conference hosted on Zoom for a selected group of journalists and business professionals, LeCun said that, for the general public, the technology used by ChatGPT may seem ground-breaking. However, what OpenAI has achieved is just a “well put together” version and not something that can be considered new for the artificial intelligence community.

“I don’t want to say it’s not rocket science, but it’s really shared, there’s no secret behind it, if you will”, LeCun told the audience of virtual attendants. He went on to explain that OpenAI’s GPT-3 technology is a combination of the work of many different parties throughout the years.

One of these pieces of the puzzle that make the AI-powered chatbot works is the use of Transformer architecture – a technology created by Google for language modeling, translation, and question answering.

Moreover, the software also uses human reinforcement to progressively enhance its responses. This means that human operators are constantly rating the responses provided by the chatbot to fine-tune it as it continues to learn from both its positive achievements and mistakes.

“So there’s a whole history of this, and this did not come out of a vacuum”, said LeCun, who is also a professor of neural sciences at New York University.

ChatGPT Empowers Both Individuals and SMBs

Even though ChatGPT may not be as innovative as the general public thinks, the software has accomplished two things that other AI platforms have not thus far. First, it has made AI a more “mainstream” topic of discussion.

Second, it has created an easily monetizable solution that both individuals and corporations can use for multiple purposes and has immediately threatened the long-standing reign of companies such as Alphabet (GOOG) by making available a powerful, and previously unavailable, tool that startups can use to build AI-powered software.

Certain businesses have already started to incorporate ChatGPT’s technology into their product offering by relying on the platform’s application programming interface (API) – a bridge that may forever change the internet as we know it.

One example is Orca Security, an Israeli cloud risk management company that has integrated ChatGPT into its risk assessment platform to help its customers remediate any weaknesses or breaches that the system identifies.

So, as LeCun asserted as well, even though ChatGPT may not be a nobel-prize winning app, it is a “well engineered” initiative that has made generative text-to-text AI more accessible to regular folks and small-to-mid-sized businesses (SMBs).

What Does the Future Hold for ChatGPT?

OpenAI – the parent company of the popular AI-powered chatbot – has received a $10 billion investment from Microsoft (MSFT) in recent days as part of a growing partnership between the two companies.

The Redmond-based firm is also the exclusive provider of cloud computing – a crucial part of the ChatGPT architecture needed to rapidly process and respond to users’ queries. Microsoft is already working on integrating the solution into some of its flagship products including the search engine Bing and its legacy world-famous productivity tool – MS Office.

Meanwhile, OpenAI could soon launch a paid version of ChatGPT that will allow users to make send an unlimited number of text prompts to the software while the company may also officially roll out the API that will allow millions of developers to connect their services to ChatGPT to take advantage of this generative AI technology.

Other Related Articles: