AI Investments Shift Towards Inference with DeepSeek and OpenAI Models

Mar 17, 2025
AI investments are expected to focus more on inference, driven by new reasoning models from DeepSeek and OpenAI, leading to increased spending by major tech companies.

DeepSeek and OpenAI are influencing a shift in AI investments towards inference, as reported by Bloomberg. The introduction of reasoning AI models by these companies is expected to increase overall AI spending, with a focus on running AI systems post-training rather than on infrastructure like data centers and chips.

Bloomberg Intelligence anticipates that hyperscale companies such as Amazon, Meta, and Microsoft will increase their AI investments faster than previously forecasted. By 2032, nearly half of all AI spending is projected to be directed towards inference, with a significant decrease in the share of investment for training.

The release of DeepSeek's models, which perform comparably to those of OpenAI and Google but at a lower cost, has prompted this shift. OpenAI's recent release of the o3-mini model, part of its o1 series, further emphasizes the trend towards reasoning models that excel in complex tasks such as science, coding, and math.

We hope you enjoyed this article

Consider subscribing to one of our newsletters like Enterprise AI Brief or Daily AI Brief.

Also, consider following us on social media:

Enterprise AI AI Brief AI Brief (X)
Free newsletter

Enterprise AI Brief

Weekly report on AI business applications, enterprise software releases, automation tools, and industry implementations.

Guide
Hosted Speech Recognition APIs

Hosted Speech Recognition APIs

Whitepaper

Tensordyne Napier: What If One Rack Could Do the Work of Nine?

Tensordyne

This Tensordyne whitepaper presents Napier, an inference-focused AI processor and rack-scale system based on the company’s TDN Math logarithmic number system. It examines infrastructure requirements for large mixture-of-experts and agentic models, compares major inference architecture approaches, and details the TDN AIP processor, TDN72 pod, TDN Link fabric, and Napier Ultra configuration. The paper reports simulation-based performance, cost, and accuracy-validation results, including Tensordyne’s projected comparison of one Napier rack with a nine-rack Nvidia Rubin plus Groq deployment; the chip is reported as taped out and in fabrication.

Read more

You may also like

Free, six days a week

Daily AI Brief: the AI news that matters, in your inbox.