After OpenAI announced the latest version of its large-language GPT model, its biggest rival in the field of generative artificial intelligence in the US announced the expansion of its own solution. Anthropic said Monday that Claude, its AI assistant, is now available in Europe and supports “multiple languages,” including French, German, Italian and Spanish in Claude.ai, an iOS app and marketing strategy for teams.
The launch comes as Anthropic expanded its API to Europe to encourage developers to use and integrate its models. Both are a part of a larger startup push to speed up growth. To date, Anthropic has raised nearly $8 billion at a valuation of $18.4 billion (post-cash), with over $7 billion raised last 12 months.
Co-founder and CEO Daniela Amodei confirmed to TechCrunch that Anthropic is in the technique of raising additional seed capital. “Yes, but we cannot comment further,” she said of the fundraiser in an emailed interview.
Anthropic’s list of nearly 60 current investors strategically includes Amazon, Google, Salesforce, SAP and Zoom. Alameda and FTX recently announced plans to sell their former shares at an increased value of $884 million in a secondary transaction related to the bankruptcy proceedings.
Anthropic is not the only company benefiting from investor appetite for backing AI startups. We know from sources close to Mistral AI, one other LLM player, that it is in talks with investors to raise almost $600 million at a valuation of $6 billion. In particular, SoftBank is not an investor in any of those corporations, which puts it among potential sponsors.
Investors are currently very enthusiastic about generative AI, but perhaps consumers are a bit less enthusiastic. As we reported last week, the Anthropic iOS app, launched in early May, has been met with lukewarm reception from users so far, underscoring larger questions about how much of the interest in artificial intelligence we’re currently seeing is just a passing fad. This may pose a challenge as the company looks for further business opportunities overseas.
Amodei believes bringing its own iOS to the OpenAI marketplace is not a straightforward comparison, given Anthropic’s primary focus on work and enterprise apps, emphasis on more “seamless” experiences that move from personal to work accounts, and switching between different interfaces and platforms, and she also felt it could have been luck in timing for the larger rival.
“ChatGPT for mobile devices came out at a time when these types of consumer apps were still very young, and a lot has changed since then,” she said. She added that “millions” of consumers in the US and UK use Claude “and we continue to see very strong adoption of our paid Claude subscription (Claude Pro) since the launch of Claude 3” – the company’s latest model, launched earlier in this 12 months.
“Our foremost focus is on work and enterprise applications – and our recent software launch Claude’s team plan indicates a continuing trend in our country. We want our users to interact with Claude in the way that feels most intuitive to them – via mobile, web or API. We’re aiming for a fairly seamless experience where Claude users can switch between personal and work accounts and switch between laptop and mobile in the same way that employees use Slack on their laptops during the workday or on their phones when they are on the way.”
Amodei declined to provide specific figures on API usage in Europe, but said it was seeing “explosive growth rates that continue to increase in key European markets such as France and Germany.” Arousing the interest of users across Europe is just one of the company’s challenges in this market.
Europe has been one of the loudest voices on AI security and regulation, especially after the passing of the Artificial Intelligence Act earlier this 12 months. Amodei believes that Anthropic is well placed to operate inside a European framework.
“Anthropic was founded on the premise of building the most secure AI systems in the industry and at the forefront of pioneering AI security research,” she said, adding that the company is working “diligently” to comply with regulations akin to GDPR in the EU. She added that there is still much work to be done on how the Artificial Intelligence Act can be implemented.
“While the Artificial Intelligence Act has been approved, there are still many steps left to develop detailed guidance on its implementation in the coming months, and we intend to work with the EU in this process.” She added that the company continues to work and contribute to industry efforts to improve AI security, including banning the use of its technology for political campaigns and lobbying, with built-in automated systems to detect related breaches and disinformation.
Her work on mechanistic interpretation – which she described as “research aimed at opening the ‘black box’ of AI models and revealing their inner workings” – reached a breakthrough in 2023 around “dictionary learning” to understand what is happening inside the AI model when it is “thinking,” she said. “We hope to ultimately use this newfound knowledge to develop methods to guide models toward safer behaviors.”
Anthropic currently employs 40 people in its London office and several contractors from European countries, Amodei said, and is preparing to hire more, particularly for the construction of the recent office in Dublin.