SAN FRANCISCO, Feb 5 (Reuters) - Technology startup Anthropic on Thursday launched what it called an improved artificial intelligence model, days after its product advances helped kick-start a selloff
Anthropic Unveils Enhanced AI Model Amidst Software Stock Decline
Anthropic's AI Innovations and Market Reactions
SAN FRANCISCO, Feb 5 (Reuters) - Technology startup Anthropic on Thursday launched what it called an improved artificial intelligence model, days after its product advances helped kick-start a selloff of traditional software stocks.
Overview of Claude Opus 4.6
The San Francisco-based lab, which is backed by Amazon.com and Alphabet's Google, said its Claude Opus 4.6 model is an upgrade to the Opus 4.5 model released in November. The new AI can work on tasks for longer and more reliably, while showing gains related to coding and finance, Anthropic said.
Market Impact and Industry Perspectives
Anthropic also teased how this tech could process 1 million pieces of data known as "tokens" in a single prompt, matching a capability earlier claimed by Google and a less powerful Claude model. And it previewed how this AI, in the computer programming tool Claude Code, could divvy up tasks among multiple autonomous agents and get work done faster.
Concerns Over Software Relevance
Seen as a disruptor in the software industry, Anthropic is aiming to stay at technology's frontier ahead of its highly anticipated initial public offering, at a time of competition from Google and OpenAI.
Connecting AI with Legacy Tools
Software developers have embraced its AI for coding. Anthropic is meanwhile making a push for business deals with products like Claude Cowork, which executes computer tasks for white-collar workers.
The AI companies' swift deployments have stoked market moves that predict older software businesses will lose relevance as AI beats them at their own game. Shares of Salesforce, Workday and Thomson Reuters each traded around 3% lower Thursday, extending declines over the past week.
Still, technology industry figures, including Nvidia CEO Jensen Huang, have dismissed such concerns about disruption, arguing that the specialized products, vast data and AI adoption of older software companies will provide a moat.
Scott White, Anthropic's head of product for enterprise, also said the goal was to connect AI to older software tools to make them more useful.
"We are excited to partner and actually lower the floor to get more value out of those tools," White told Reuters.
Claude Cowork, he said, is more like "the front door to getting hard work done."
(Reporting by Jeffrey Dastin in San Francisco; Editing by Lisa Shumaker)
