Tech Desk 
Jul 13, 2024
OpenAI has rapidly ascended to new heights since its launch, attracting millions of subscribers worldwide. The company has achieved numerous milestones but aims to attain Artificial General Intelligence (AGI), aspiring to become an AI superintelligence.
According to Bloomberg, OpenAI has outlined five key steps to achieve AGI, which are summarized below:
Levels to reach AGI:
1. Chatbots
The initial level towards AGI involves "Chatbots" or AI capable of conversational language with enhanced natural tones that are easy to comprehend. Models like GPT-4o, Gemini Pro 1.5, and Claude Sonnet 3.5 already excel at complex conversations.
2. Reasoners
The next level, termed "Reasoners," aims for human-level problem-solving across a broad spectrum of areas without specific task prompts. OpenAI plans to release GPT-4.5 later this year with significant reasoning enhancements.
Read more: OpenAI gets supercharged — Accessing the new GPT-4o capabilities
3. Agents
At the third level, AI becomes capable of performing tasks autonomously without human input or detailed instructions. OpenAI CEO Sam Altman hinted at the upcoming release of GPT-5, an agent-based AI platform.
4. Innovators
Moving forward, AI evolves into an innovative platform capable of aiding in invention and generating creative content autonomously. For example, it could create new languages from scratch without explicit instructions.
5. Organisations
The final stage envisions AI efficiently managing organizations independently, operating on its own terms without human or third-party assistance.
OpenAI anticipates achieving AGI by 2028, positioning itself at the forefront of AI development in the coming decade, according to Altman.
