Hemanth Mandapati, boss of German start-up Novo AI, was an early adopter of DeepSeek technology when he switched to the Chinese firm’s artificial intelligence (AI) model from ChatGPT creator OpenAI’s own model two weeks ago.
“If you have built your application using OpenAI, you can easily migrate to the other ones … it took us minutes to switch,” he said in an interview on the sidelines of the GoWest conference for venture capitalists in Gothenburg, Sweden.
DeepSeek’s emergence is changing the landscape for AI, as it offers companies access to advanced technology at a fraction of the cost, according to interviews with more than a dozen start-up executives and investors. It also has the potential to push other AI companies to improve their models and bring down prices.
“There was an offer from DeepSeek which was five times lower than their actual prices,” Mandapati said. “I am saving a lot of money and users don’t see any kind of a difference.”
Europe’s tech start-ups had struggled to adopt AI models at the same rate as US rivals, which have easier access to funding. But executives said DeepSeek could be a game changer.
“It marks a significant step forward in democratising AI and levelling the playing field with Big Tech [firms],” said Seena Rejal, chief commercial officer of British firm NetMind.AI, another early adopter of DeepSeek.
Analysts at Bernstein estimate that DeepSeek’s pricing is 20 to 40 times cheaper than equivalent models from OpenAI.
OpenAI charges US$2.50 for 1 million input tokens, or units of data processed by an AI model, while DeepSeek is currently charging US$0.014 for the same number of tokens.
Concerns have been raised by regulators about whether DeepSeek is copying OpenAI data or censoring answers that could portray China in a bad light. It is currently being investigated in different European countries.
“While the future of DeepSeek as a business is difficult to predict, the structural impact seems quite pervasive,” said Sanjot Malhi, partner at venture capital firm Northzone.
Nearly US$100 billion was invested last year by venture capitalists in AI companies in the US, compared with about US$15.8 billion in Europe, according to data from PitchBook.
Just on January 22, the Trump administration unveiled a US$500 billion AI infrastructure project called Stargate, a joint venture backed by OpenAI, SoftBank Group Corp and Oracle.
Meanwhile, AI investments in Europe have been more modest. Only France’s Mistral features among the list of top foundational models dominated by the likes of OpenAI, Meta Platforms, Anthropic and Google.
China’s DeepSeek attracted global attention after publishing a technical paper last month, which said the training of its namesake V3 AI model required less than US$6 million worth of computing power from Nvidia’s H800 chips. DeepSeek’s own chatbot has since overtaken ChatGPT to become the top-rated productivity application available on Apple’s App Store.
“This is a wake-up call that bigger isn’t always better,” said Fabrizio Del Maffeo, chief executive of Axelera AI. “By making models more attainable to everyone, the total cost of ownership and barriers to building innovative tech are lowered, which can be a catalyst for the whole industry.”
While some analysts doubt that DeepSeek’s training cost is as low as the company touted, they agree it is lower than the amount spent on comparable American AI models.
“I see DeepSeek as a tremendous opportunity for companies like ours,” said Ulrik R-T, chief executive of Denmark’s Empatik AI. “It showed that we do not need huge budgets to be able to achieve our vision.”
But a price war may have already started. Microsoft last week released OpenAI’s o1 reasoning model to all Copilot users for free, instead of the usual subscription fee of US$20 per month.
“AI prices are going down, so future usage is probably going where there is transparency, which is usually open source, even though it’s in China,” said Scale Capital’s Joachim Schelde.
Bigger companies, ranging from Finland’s Nokia to Germany’s SAP, are more cautious about switching.
“Cost is just one factor,” said Alexandru Voica, head of corporate at Britain’s Synthesia, which was last valued at US$2.1 billion. “Other factors are: ‘do you have all the security certifications, the frameworks, the software ecosystem that allows companies to build and integrate with your platform?’”
Source : Reuters
Bd-pratidin English/ Afia
