In the last few years, Artificial Intelligence has seen an upward growth across all sectors. In such a short span of time, AI has been talked about so widely that now, if a technology is not using AI it doesn’t seem worthy enough. As AI systems are becoming more sophisticated, the need to measure their intelligence is becoming greater. Recently, OpenAI has been thinking the same. So, the AI ​​research company came up with an internal scale to track the progress of its big-language models towards Artificial General Intelligence. According to a recent report, the company has created this scale to determine how smart its AI tools are.

A Bloomberg The report states that according to this scale, OpenAI’s ChatGPT is at Level 1. To measure the progress, OpenAI created this scale with level-wise separation. It has also defined every level and how it will be measured. It is interesting to note that these levels are compared to the human ability to solve a task.

OpenAI measures the smartness of AI

The AI ​​Intelligence Framework defines five levels of sophistication. OpenAI claims that ChatGPT exhibits Level 1 qualities, but it will move closer to Level 2 very soon. Level 2 AI corresponds to the problem-solving abilities of a PhD holder, while Level 3 AI can act autonomously on behalf of the user. Level 4 AI exhibits creative abilities, generating new innovations.

Finally, Level 5 AI, which is the pinnacle of achievement, can replicate the collective efforts of an entire organization, effectively achieving Artificial General Intelligence (AGI). OpenAI has previously defined AGI as “a highly autonomous system that surpasses humans in the most economically valuable tasks.”

The level system shows that OpenAI is aiming to achieve AGI level intelligence through its tools. A few months ago, OpenAI CEO Sam Altman made headlines by saying that he wanted to create AGI (artificial general intelligence), no matter what the cost.

OpenAI also noted, without providing further details, that the GPT4AI model offers new skills that exhibit human-like reasoning. Bloomberg report. This internal scale comes after OpenAI’s CEO hinted about the company’s next LLM, GPT-5. According to him, GPT-5 will be a “significant leap forward”.

OpenAI ChatGPT 5

According to a report DecoderAltman spoke at the Aspen Ideas Festival and expressed optimism about GPT-5. However, he also said that there is still a lot of work to be done on the new model. Despite this, he expressed confidence in its potential and said that GPT-5 aims to correct many of the errors seen in GPT-4.

While there is no clarity about the new ChatGPT-5, Altman’s comments suggest that the model is still in the early stages of development. The hype around GPT-5 follows the impressive debut of GPT-4o, which demonstrated its mathematical prowess and ability to engage in emotionally intelligent conversations, raising the bar for its successor and increasing anticipation of even more unprecedented capabilities. OpenAI hosted its Spring Update event in May this year and introduced GPT-4o to the world.