OpenAI's ambitious AI project, GPT-5, codenamed Orion, is significantly behind schedule and grappling with uncertainty regarding its effectiveness, as reported by the Wall Street Journal.

Initially expected to launch in mid-2024, the project has faced numerous setbacks, including issues with the quality and diversity of training data.

A primary challenge has been the realization that there is insufficient public internet data available for effective training, which differs from the resources available for previous models.

Despite extensive training runs, Orion has not justified its substantial costs, which are estimated to exceed $500 million for a six-month training period.

Meanwhile, Elon Musk's xAI is expanding its Colossus supercomputer in Memphis, Tennessee, aiming to enhance its GPU capacity significantly.

Experts believe that if successful, GPT-5 could revolutionize various industries by improving customer service, digital marketing, and operations like contract analysis.

The Wall Street Journal's findings align with earlier reports indicating that OpenAI is adjusting its strategies in response to a slowdown in AI improvement rates.

Technical difficulties have plagued the project, with at least two major training runs failing to yield the anticipated results.

To address data shortages, OpenAI is employing software engineers and mathematicians to generate new content, a process that is both slow and labor-intensive.

Compounding these challenges, OpenAI has experienced significant internal turmoil, including the departure of over two dozen key executives this year due to competitive offers from rival companies.

As OpenAI struggles with Orion, competitors are gaining ground, with some releasing models that industry experts rate as superior to GPT-4.

In light of OpenAI's slowdown, Amazon has invested an additional $4 billion into rival Anthropic, raising its total investment to $8 billion.