OpenAI Takes a Bold Step Toward Artificial General Intelligence

Two years ago, tech startup OpenAI introduced its innovative chatbot, ChatGPT. This launch is often viewed as the start of the Age of AI.

However, that impressive debut may have just been the beginning for OpenAI.

The company appears to be gearing up for its next major chapter, which could surpass its initial success and position OpenAI as a leading force in the global tech landscape.

The firm is making significant strides toward achieving artificial general intelligence (AGI), seen as the ultimate goal in AI development.

AGI represents the highest level of AI capability. According to McKinsey & Company, AGI will possess “human-like cognitive abilities, including reasoning, problem solving, perception, learning, and language comprehension.”

“AGI tools may exhibit cognitive and emotional skills (like empathy) comparable to human abilities. Depending on one’s definition of AGI, they may even possess a conscious understanding of their actions.”

Once this AI breakthrough is reached, it could significantly alter our daily lives.

Currently, OpenAI is making impressive progress toward that extraordinary goal.

Progressing Towards Artificial General Intelligence

Recently, the company unveiled a new advanced AI model – ChatGPT o1 – designed for complex reasoning tasks.

Many AI models exist today, yet most share similar structures and capabilities. Models like ChatGPT 4, Gemini, Claude, and Grok primarily function as conversational agents with limited functionality.

These chatbots can conduct conversations, respond to inquiries, and complete various tasks. However, they mainly rely on pre-set scripts or basic pattern recognition.

In essence, they do not “think” in the way humans do.

ChatGPT o1, however, represents a significant advancement in this field.

Going beyond simple recognition, ChatGPT o1 demonstrates logical reasoning, the ability to infer information, comprehend cause-and-effect, and solve problems using principles learned during its training. This capability allows it to assist not just in automating tasks but in tackling complex problems.

During the introduction of this model, OpenAI stated, “o1 can be used by healthcare researchers to annotate cell sequencing data, by physicists to generate complex mathematical equations for quantum optics, and by developers across various industries to create and manage multi-step workflows.”

This notable advancement marks a pivotal moment in AI evolution, producing remarkable results.

Each week, the TrackingAI project administers IQ quizzes to multiple AI models. Most current models, such as ChatGPT 4, Llama 3.1, Claude, Grok, and Bing Copilot, generally score between 80 and 90 on these tests.

In contrast, the average human IQ is around 100. Thus, existing AI models are slightly less “intelligent” than humans on average.

In a remarkable twist, ChatGPT o1 scored 124 on the TrackingAI IQ test.

This score exceeds current AI model averages by nearly 50% and nears the genius threshold, which is an IQ above 130, representing just 2% of the human population.

ChatGPT o1’s breakthrough in complex reasoning is a significant leap in AI advancement.

Conclusion

This groundbreaking achievement brings OpenAI closer to realizing AGI.

It may also explain why the company is seeking to raise billions of dollars at a staggering valuation of $150 billion, drawing interest from key investors.

Microsoft, Apple, and Nvidia are reportedly among potential investors. The alignment of these tech giants investing in the same startup is unprecedented.

When was the last time Microsoft, Apple, and Nvidia all invested in one startup? It remains an unanswered question, highlighting the unique nature of this situation.

We even posed this inquiry to ChatGPT, acknowledging its potential biases in responding.