Blackstone reportedly increases Anthropic stake to $1b

Blackstone has increased its investment in AI startup Anthropic, raising its stake to about US$1 billion, according to sources familiar with the matter.

The firm is investing an additional US$200 million in the ongoing funding round, valuing Anthropic at around US$350 billion.

Anthropic, backed by Amazon and Alphabet, develops AI models such as the Claude family and recently launched a new system called Opus 4.6, which offers improvements in reasoning, coding, and complex text generation.

The company claims the new model provides better performance in tasks like software development and financial analysis, with enhanced reliability and longer operational periods.

The investment comes amid heightened investor interest in generative AI companies and follows recent drops in software stocks, driven by concerns over AI’s potential to disrupt traditional business models.

Neither Blackstone nor Anthropic immediately responded to requests for comment.

🔗 Source: Reuters

🧠 Food for thought

Implications, context, and why it matters.

This valuation is fueled by a capital arms race

Blackstone’s investment is part of a funding round that could top $20 billion, up from an early $10 billion goal, after strong investor interest, according to people familiar with the matter 1.
The push comes five months after Anthropic raised $13 billion at a $183 billion post-money valuation 2.
Costs keep climbing as Anthropic races OpenAI and pays for computing power plus scarce talent to build leading AI models such as Claude and Opus 4.6. Nvidia (a leading AI chipmaker) and Microsoft are expected to supply much of the money, which ties the AI ecosystem closer and raises the cost of competing 3 1.

Anthropic is pitching Opus 4.6 as a tool for enterprise knowledge work

Opus 4.6 targets business customers for reasoning, coding, and complex text generation. The release landed within an hour of an OpenAI update, which tightens the fight for enterprise AI workloads 4.
Pricing is $5 per million input tokens and $25 per million output tokens. DeepLearning.AI (an AI education and analysis platform) called that about 60% cheaper than comparable flagship models while citing Anthropic’s benchmark claims 5.
Lower prices for top models could squeeze enterprise software vendors and data providers as executives weigh AI for financial analysis or software development. Anthropic’s recent legal and business research models also shook the share prices of publicly traded data firms, according to TechCrunch’s summary of the Bloomberg report 3.