When you buy through our articles links, Future and its syndication partners can earn commissions.
Credit: Justin Sullivan via Getty Images
Sam Altman’s predisposition to Opeenai CEO, say, to be investigated, but not to find out that it is one of the highest technology organizations on the planet, we will say that it is impressive. Sometimes he doesn’t seem to really think before speaking. The latest example includes his status as a “New Father”, which he thinks he does not seem to consider the prospects without the help of his own conversation program (through Techcrunch).
“Obviously, people have been able to take care of babies for a long time without conversations,” says “I don’t know how I would have done it.”
“Those first few weeks were all the time,” he says about his tendency to consult the CHATGPT on childcare. Apparently, books, counseling friends and family members, even old -fashioned Google Search for this Colossus would not have executed artificial, ER, Intelligence;
If all this is a touch arch, forgive me. However, Altman operates in this interview in absolutely AI evangelization mode. “I spend a lot of time thinking about how my child will benefit in the future,” he says, “my children will never be smarter than Ai. But they will grow much more capable than we grew up and will be able to do what we can’t imagine, they will really use AI. “
There are many urgent and obvious contradictions about this worldview. Of course, people will be better off using AI. But will they be more capable? Perhaps most people will not be able to write a consistent prose if he is doing it from day one. Will Ai write everything that will be capable?
Not that this is the main revelation, but this podcast makes it possible to understand how the altman is signed for the Ai revolution. “They will look at it as a very prehistoric period,” he says about today’s children.
This is a somewhat strange statement, taking into account the “prehistory” before the human activities and efforts were recorded, because of the offspring. And, of course, the existence of large language models created by “Open” depends entirely on the large number of data gigabytes that originally taught those LLM.
In fact, one of the biggest challenges currently facing AI is the concept of Chatbot contamination. The idea is that since 2022 With the release of the Chatgpt into the wildlife, the data that are now taught to the LLMS increases contaminated with synthetic output of previous conversation data.
As more and more conversation data on the common pool inject more and more synthetic data, later generations of AI models will become increasingly contaminated and less reliable, which will eventually become a state known as the collapse of the model.
In fact, some observers believe that this is already happening, as evidenced by the increasing predisposition of some of the latest models to hallucinations. Cleaning this problem in some accounts will be “extremely expensive, probably impossible”.
In any case, if there are problems with Altman’s infinitely optimistic sayings, it is probably a lack of nuances. Everything is desperate and rude, until it is difficult to imagine how you would look after the newborn baby without conversations. Everything beneath ai is bright, clean and perfect.
Of course, anyone who has used the current chat program for more than a few moments will be very familiar with their immediately obvious restrictions, not to mention the broader problems they can cause, even if problems such as hallucinations are overcome. At least it would be much easier to empathize with similar altmen if there was a bit of a feeling that these challenges would balance his one -sided narrative.
Anyone, release the podcast and decide for yourself what you do from the attitude of the Altman’s All-A.