Musk turns to rival Anthropic to power Grok
Elon Musk has opted to buy competing technology as he is set to use rival Anthropic’s AI models to power parts of Grok, despite spending heavily to build his own AI challenger.
Musk said on Wednesday that the Grok bot would use the “best back-end model for any given task”, including Anthropic’s latest Claude Opus chatbot, alongside image generator Midjourney and AI music platform Suno.
“Whatever is most likely to give you the best outcome”, Musk wrote on X.
The move is a notable departure for Musk, who has poured billions of dollars into building xAI and Grok as a rival to OpenAI and Anthropic. Rather than relying exclusively on its own technology, the Grok bot can now route work to whichever outside model it judges best suited to a particular request.
It also gives Anthropic another commercial boost as the AI company takes an increasingly large share of business spending ahead of a potential blockbuster stock market listing.
Data published by City AM last month showed Anthropic accounted for as much as half of UK businesses’ spending with major AI providers in September, based on Capital on Tap transactions from more than 300,000 British small businesses.
ChatGPT’s share fell from 60 per cent to 36 per cent, while more than one in seven businesses paid a major AI provider during the month.
Musk shops around for AI
Musk launched xAI in 2023 as a direct challenger to OpenAI, which he co-founded before leaving, and has invested heavily in the computing infrastructure needed to train its Grok models.
The AI operation has since been brought under SpaceX, while SpaceXAI launched Grok Bot in August as a system of individual agents equipped with their own computers and capable of completing several tasks in parallel.
Wednesday’s announcement means those agents will not necessarily use Musk’s own underlying models to do that work.
The move could signal a broader shift in how AI companies compete. As models become better at different tasks, owning the interface that chooses between them could prove as commercially important as building the underlying technology itself.
It follows disruption to Grok on Tuesday, when users reported problems accessing the chatbot across its mobile and web services.
More than half of complaints tracked during the disruption related to AI generation, although no official cause was given.
Anthropic, meanwhile, is preparing for a potential stock market debut that its investors expect could value the company at $2tn or more.
OpenAI has delayed its own potential listing as it works through a series of safety concerns.