PARMY OLSON: Zuckerberg introduced artificial intelligence for free

There is a mystery around Mark Zuckerberg, not his new necklace. It is that the CEO of Meta Platfarms has spent billions of dollars to build strong artificial intelligence models and offer this technology for free. So why is he doing this? Zuckerberg recently claimed that its artificial intelligence model for training is “Llama” open source and the concentration of power is in the grip of one business. Doubt and Dariah was the president of “Mita” in the Podcast’s “Morning Pro Daily” newspaper in February and says, “I really focus on open sources. My theory of this is that what you want to appear is similar to a more advanced and powerful setting than others.” Perhaps this interpretation should be seen with suspicion, as it is eventually issued by a man who has succeeded in strengthening his dominance on social media by buying and mimicking competitive platforms, and a semi -thickness control over ‘death’ with more than 50% of the total vote. Like any emperor who loves goodness in the ‘Silicon Valley’, Zuckerberg gave a character to guide the public advantage for his efforts to acquire his moral value of his business. The real reason for his artificial intelligence shared is due to an effort to make things more difficult for its competitors, as well as to improve the ‘dead’ reputation, so that you can attract the most experienced engineers for artificial intelligence. It can also be allowed to explore new ways to improve its advertising work. In light of this, investors must rejoice, but the developers of artificial intelligence with human ideals must think twice before joining it. An unintentional bet at this point requires an unintended bet on the side of Zuckerberg. At the end of 2022, before the launch of the ‘Openai’ Chat ‘GBT-4’, the company ‘Inviteia’ is given a large negative version of the purchase of graphic processing units-the strong chips used to build most of the artificial intelligence systems according to an interview with the ‘CARE’ website that specializes in technological journalism. It was aimed at improving the recommendation system (a branch with information filtration systems aimed at expecting the user’s admiration for a specific product) to serve the short videos (roles), which is the reproduction of ‘dead’ for the short videos website “Tech Talk”, but it appears that Zuckerberg is unintentional in the world of technology these days, Artificial intelligence chips that showcased the exhibited. Currently. This is one of the reasons for “Mita” that has been very quickly in the field of artificial intelligence since then, as it introduced its first version of “Lama” in February 2023, and then “Lama 2” after only five months. When Lama 3 was launched last month, the system immediately entered the highest global rating of artificial intelligence models (currently sixth), making it the most advanced system that each can use for free. Trouble with competitors, “Lama”, which is the system that also operates the new artificial intelligence assistants of “Mita”, makes things free for senior competitors such as “Microsoft” and “Amazon.com” that all receive fees for their artificial intelligence services within their business in the field of wool computer. This step is supposed to take a long way to attract more distinctive talents, which is a very important issue for Zuckerberg, to the extent that he personally contacted the email researchers with the “Deepmind” business, “Google”, to provide them with job offers, according to the “Informination” website. Many artificial intelligence developers want a fixed salary and an opportunity to have a widespread impact. The Lama version has so far been downloaded more than 100 million times, and software developers want to achieve its ‘weights’, the internal institutions that help him process data, which makes the model more adaptable to businesses that want to summarize legal documents or classify customer commentary. This flexible technology construction approach also makes Mita a more attractive place to work. “Google” simulation? In the future, there could be another theoretical benefit of Zuckerberg’s great December in the field of artificial intelligence: Google’s success simulation with the Android system. The “Alphabet” subsidiary began providing its cellphone operating system to the hardware manufacturers in 2008, and over time it added requirements that include applications such as “Google Serch”, “Google Maps” and “G -Mail”. As the number of Android devices increased, the demand for these services increased, which enabled Google to earn revenue through advertising. Google applications on the Android system are mainly gates for billions of user and a new income. Is “death” benefit from “Lama”? In theory, and after the model has become more common between developers and an integral part of their applications, “dead” can add a condition for these users to share the data they process, which can be used to improve the company’s artificial intelligence models, and even improve the advertising activity. By doing so, this means, of course, that Zuckerberg can no longer describe “Lama” as “Open Source”, but there is already controversy among industry groups about whether the model is considered in the first place. This has already happened, as “Mita” wants to get more control on how to use each new version of “Lama”. The Lama 3 License Agreement seeks to provide more protection for its intellectual ownership, and a greater compliance with its commercial sign and legal standards, for example, more than the “Lama 2” agreement. An apparent transformation will not be such an unknown transformation. You just have to look at the company “Oben Ai”, which was originally used as “open” term for “employment purposes”, and over time it has more secret and limited. Two senior researchers in Open Source Artificial Intelligence told me that Mita will be able to set provisions for the part of data in the future licensing agreements for “Lama”, if the company limits a few open sources of the model. And if “Mita” can acquire large amounts of unknown data that created “llama”, it can help her improve her understanding of the content that attracts users and improve how to assign her ads (to direct certain ads to certain people according to their preferences). It will not all appeal to. Fear of collecting personal data was the reason behind the violation of the Italian privacy laws. But Zuckerberg turned ‘Facebook’ into a money printer after dominating the social media market and gathering hundreds of millions of dedicated users. It prefers a strategy to first track user preferences and then gain revenue later. And if he does the same with “Lama”, it could mean that the artificial intelligence model is also less open and more profitable for his business.