The artificial intelligence sector continues to advance by leaps and bounds, but the pace has not been as sustained as many expected. OpenAI, known for its revolutionary ChatGPT, is at the center of this scenario with its latest model, called Orion.
Although the company had high expectations for this AI, Orion’s performance has not reached the expected level. In particular, he has shown limitations on complex tasks, such as answering coding questions outside his training area. This result marks a departure from the significant advances made in previous iterations, such as the jump between GPT-3.5 and GPT-4.
Orion and the scarcity of training data
One of the biggest challenges in the development of Orion has been the scarcity of high-quality data for its training, especially in the field of coding. As artificial intelligence advances, access to fresh, useful data sources has become increasingly limited.
OpenAI, like other big tech companies like Google and Anthropic, faces a crossroads: improving AI models without relying on large amounts of data and keeping huge processing and development costs under control. This raises questions about the viability of continued investment in models that do not necessarily offer a return proportional to the resources used.
The difficulty of scaling AI
The unofficial law of scaling in AI, which assumes that more data and greater processing power will always lead to better results, is beginning to find its limits.
Margaret Mitchell, an AI ethics scientist at Hugging Face, noted that the industry is realizing that “scaling laws are not absolutes.” According to Mitchell, achieving advanced artificial intelligence that performs at a human level could require “different training approaches.” This outlook reflects uncertainty about whether the Orion model, and other similar developments, will be able to achieve the performance that consumers expect from leading companies.
Orion’s Influence on the AI Market
In this context of high expectations and uncertain performance, OpenAI has not been the only company facing these obstacles. Google, for example, has had problems with the Gemini model, while Anthropic has seen delays in the release of its Claude 3.5 Opus model.
These difficulties also raise questions about the ability to achieve Artificial General Intelligence (AGI) in the coming years, a goal that refers to a type of AI capable of matching or surpassing human performance in a wide range of intellectual tasks.
What is the way forward?
To overcome these challenges, OpenAI and its competitors are exploring new “reasoning” techniques in AI. The idea behind these advances is that future models will be able to process complex queries and make decisions based on multiple options before providing a final answer.
Sam Altman, CEO of OpenAI, recently mentioned that although the company is still working on improvements for models like Orion, the next big advance will come from the implementation of “AI agents” that will be able to perform automated tasks, such as booking flights or sending emails.
Final thoughts
The development of artificial intelligence is facing a crucial moment in which technical challenges, lack of quality data and high costs are forcing companies to reconsider their strategies. Attention has begun to shift from building larger and larger models to creating specific applications that offer added value to users.
The AI industry could be on the threshold of a new stage, where efficiency and specialization become key priorities for success in a rapidly growing sector.
Do you want to always be updated in the world of cryptocurrencies? Subscribe now to the CriptoTendencia WhatsApp channel! Here you will instantly receive the most relevant information about Bitcoin, Altcoins, DeFi, NFTs, Blockchain and the Metaverse.
Related
Crypto Keynote USA
For the Latest Crypto News, Follow ©KeynoteUSA on Twitter Or Google News.
