There is blood in the water, and the sharks have smelled it.

There is blood in the water, and the sharks have smelled it. The AI price war has begun.

I'm going to be honest, with the dominance OpenAI and Anthropic have shown in the frontier AI model space (and Google hanging in there), I thought the other tech giants were cutting their losses and moving on.

Anthropic has been very clear that their strategy is to target enterprise to reach profitability. To reach that profitability, we’ve seen a trend of increasingly more expensive AI solutions, whether it’s a model like Claude Fable, or a product like Claude Tag, which is very token hungry.

But out of nowhere, here comes Grok 4.5. Claiming to use 4.2x less tokens than Claude Opus 4.8 on similar tasks. Not only that, Grok 4.5 costs 1/2 of Opus for input tokens, and 1/4 of Opus for output tokens.

Just one day later, after reports of Meta trying to sell off their AI infrastructure, and seemed to be pivoting away from their own frontier cloud AI models, Muse Spark 1.1 was announced. 1/4 the price of Opus for input tokens, and almost 1/6 the price of Opus for output tokens.

It seems like both SpaceXAI and Meta found a chink in Anthropic’s armor.

Competition is great for the industry. With the sudden shift from tokenmaxxing (awful term) to tokenomics, and with companies under pressure to have an ROI on their exorbitant AI budgets, if you want to get AI market share, now is the time to do it.

If you’re a business leader and want to know how you take advantage of multiple models in your company’s AI strategy, and prevent vendor lock-in, DM me!