News

xAI’s Next-Generation AI Model Didn’t Arrive On Time

Last summer billionaire Elon Musk the founder and CEO of the AI company xAI address Grok 3 as the xAI’s next major AI model. That would arrive by the “end of year” 2024. Grok, xAI’s answer to the models are like Open AI’s GPT-4o and Google’s Gemini, and analysis of the images and responses and the queries and the powers a number of features on X. We get this all information from his social media accounts.

Musk also wrote in a post on X referring as 

“Grok 3 end of year after training on 100k H100s should be really something special,”

“Grok 3 will be a major leap forward,” he added this statement in the mid-December follow-up post.

Yet January 2, and Grok 3 haven’t arrived and not these are the signs that are rollout. Some of the code on the “xAI’s website spotted by AI tipster Tibor Blaho suggests that an intermediate model, “Grok 2.5,”

This was also granted as the first time when Musk set a lofty goal and missed it. This is well established that Muisk’s pronounced about the timing of the product launches often works as unrealistic. In an interview with podcaster Lex Fridman in August, Musk addressed the “Grok 3 would hopefully be available in 2024 if we’re lucky.”

However, Grok 3’s MIA status is intriguing since it fits into an expanding pattern.

Anthropic, an AI startup, was unable to provide a replacement for its flagship Claude 3 Opus model last year. Anthropic extracted all references to the Claude 3.5 Opus model from their developer documentation a few months ago. They did it right after declaring that a next-generation model would be available by the end of 2024. (One report claims that Anthropic completed training Claude 3.5 Opus at some point last year. They concluded that it wasn’t financially viable to release it.)

It has been reported that in recent months, Google and OpenAI have also had problems with their flagship models.

This is proof of the shortcomings of the existing AI scaling laws. These are the strategies businesses are employing to boost their models’ capabilities. In the not-too-distant past, training models with ever-increasing data sets and vast quantities of processing power allowed for significant performance gains. However, the benefits have started to decline with each model generation. That also prompted businesses to look for different approaches.

Musk himself hinted at this in the Fridman interview.

Fridman asked in an interview

“You’re hoping for [Grok 3] to be state-of-the-art?”.

And Musk’s reply is 

“Hopefully,” “I mean, this is the goal. We may fail at this goal. That’s the aspiration.”

There could be other reasons for Grok 3’s delay. xAI has a much smaller team than many of its rivals, for one. Nonetheless, the slipped launch time frame adds to the body of evidence that conventional AI training approaches are running up against a wall.

Read Also:

  1. Accel Closes A New India Fund Worth $650 Million
  2. OpenAI Failing To Supply The Opt-Out Tool Which Is Promised By 2025
  3. Cybertruck blast right outside the Trump hotel in Vegas leaves 1 dead, and 7 injured

News Source

Avijit Sah

Avijit Sah is a digital marketing expert specializing in SEO, social media, and content strategy. With a passion for helping businesses grow online, Avijit Sah uses data-driven tactics to boost visibility and engagement. Follow Avijit for the latest digital marketing tips and insights.

Related Articles

Leave a Reply

Your email address will not be published. Required fields are marked *

Back to top button