- India
- International
Ever since the launch of GPT-4, a lot is being written about its astounding capabilities. Perhaps, its precision and accuracy have also triggered the worries of many who have been increasingly wary of the rapid developments in AI technologies. Consequently, there has been a lot of chatter around the forthcoming advanced version, GPT-5.
While some reports suggested that OpenAI is planning to complete the training of its latest iteration by the end of this year, some claimed that the GPT-5 will be pathbreaking as it will be able to achieve Artificial General Intelligence (AGI), a stage where machines will be able to display human-like intelligence and comprehension.
Now, it seems GPT-5 will not be seeing the daylight anytime soon. OpenAI CEO Sam Altman has finally broken his silence on the recent open letter that sought a monetary halt on the developments of AI systems. Altman, while speaking at an event at MIT, shared his views on the letter. He also stated that his company was not working on GPT-5 for now. He asserted that OpenAI was not training GPT-5 and would not do so for some time.
Talking about the open letter, Altman said that he agreed with parts of it. He said that OpenAI spent over six months training GPT-4 before releasing it to study the safety and get external audits and “red teamers” to understand the process and mitigate issues. He also said that ever since the launch of the GPT board, many have hailed it as the most capable model and by far the safest and most aligned one.
The Stanford University dropout said that he agreed that there was a need to get more serious about the safety capabilities and that the safety bar has to be increased. However, he also lamented the lack of technical nuance in the letter, especially on where exactly the pause was needed in the developments. He said that an earlier version of the letter claimed that his company was training GPT-5, a claim which he refuted.
“I think moving with caution and increasing rigour for safety issues is really important. The letter, I don’t think, is the optimal way to address it,” Altman said in his virtual address at the event.
Talking about the philosophy of OpenAI, Alman said that AI is going to impact all; hence it was important to engage with everyone. “Our goal at OpenAI is to get the world to engage with this and think about it and gradually update and build new institutions or adapt our existing institutions to be able to figure out what the future we all want,” he said.
A few weeks ago, about 1000 dignitaries, including billionaire Elon Musk and Apple co-founder Steve Wozniak, signed an open letter seeking a six-month halt on developments on AI systems that can compete with human intelligence.
The letter, which until a few weeks ago had over 13,000 signatures, expressed apprehension on the development of programs such as OpenAI’s ChatGPT, Bing AI Chatbot, and Google Bard as they may have negative consequences if left unchecked, leading to widespread disinformation and robbing human jobs.