By Charudutta Panigrahi*

Deep Seek hurtles us to back to the drawing board. We are hit and shaken by a not for profit, research organization born recently in 2023 puncturing the hegemony of tech lords. Technology is not an equalizer, leveler only but innately democratic. Only that sometimes political mean-spiritedness relegates it but cannot for long. Stock market jolts and valuation illusions lay bare and remind us of the all-encompassing ephemerality. Deep Seek’s AI model, which is at par with OpenAI’s ChatGPT, has costed less than $6m (£4.8m). Microsoft, for OpenAI, plans to invest about $80bn this year. 6M:80B is the ratio at which the productivity, efficiency amy be compared. Simply put it is approximately 1:13,333. Now does it surprise you that on 27 January, Nvidia lost about 17%, about $593 billion in market value, which is a record one-day loss for any company. Not only that. The shares of semiconductor companies, and other companies (like power and infrastructure) using AI, collectively had more than $1 trillion decamped from their worth. While we convalescence, we get a little philosophical and take refuse in Jevon’s paradox, self-sermonizing the greater good for AI’s increased use due to enhanced accessibility and affordability. Without Deep Seek, we would have been swimming under the tides of billions worth AI at Silicon Valley.

The timing of the announcement and the underdog story behind is bang on. Liang Wenfeng, the millennial hedge fund entrepreneur and the boss of Deep Seek says that “Our biggest challenge has never been money, it is the embargo on high-end chips,” In 2022 Biden’s announcement barred China from accessing the Nvidia H100 chips necessary for rapid AI development. Instead, low power H800 chips were developed for the Chinese market. But this was banned in 2023. On Trump’s taking on the reins, the subtle message conveyed by China is one of technology prowess, tenacity, ingenuity and the consistency of Chinese disruptive innovations keeping costs the unwavering benchmark. Foreign policy has "forced Chinese companies like Deep Seek to innovate" and Deep Seek has bolstered a sense of nationalism and now the belief that they can do more with less. Reinforcing a David Goliath retake except that here David is from world’s second largest economy not so unarmed. The same Monday, 20 Jan, Liang and the Chinese Premier Li Qiang met and the Chinese media described Liang as a "technical idealist". The Chinese state and Chinese R&D eco-system are seldom sundered. State endorsement comes fast and encompassing. Liang wants to keep Deep Seek as an open-source platform and this gives AI a more humanitarian character that could allow young start-ups to pool resources and advance faster. This brings back the focus on non profit research space and its contributions to cost solutions. The discussions now hover around how Washington’s ban was redirected and the challenges converted to opportunities for the Chinese AI industry. Deep Seek is poised to change the fundamentals of the game.

This is almost an affront to the tech czars of Silicon Valley because the rules of the AI space is being set by Deep Seek and more Deep Seeks of tomorrow from across. This has much bigger connotations. President Donald Trump responded to this development and remarked that “the sudden rise of the Chinese artificial intelligence app DeepSeek “should be a wake-up call” for America’s tech companies. The blockbuster success of yet another Chinese app came as a solid challenge to the administration and the congress. The knock is palpable and so is the paramount role of R&D & technology in geo-politics. Labs will rule.

This competitiveness will be good for the world, its consumption levels and its growth in these VUCA (volatile, uncertain, complex, and ambiguous) times. Trump said that he considered the low-cost model to be “very much a positive development” for AI overall, because “instead of spending billions and billions, you’ll spend less, and you’ll come up with, hopefully, the same solution.”

India has ties with both the United States and China and is doing a pragmatic balancing act to cope, withstand, and shape the dynamics. India is making moves to spruce up its local manufacturing capabilities to seize the ‘China-plus-one’ opportunity. At the same time India is unlocking land, power and infrastructure knots (demystifying Gordian Knots) to compete with the countries already in the reckoning. Though a NITI Aayog report, mentions that India has seen “limited success so far” in capturing the ‘China Plus One’ strategy, efforts are on, and the CEO Mr Subrahmanyam has suggested that forthcoming trade policies under US President-elect Donald Trump could augur well for India. Mr Panagariya, Chairman of the 16th Finance Commission, is of the opinion that “barring few sectors, India can open up to investments from China”. NITI Aayog vice chairman Suman Bery emphasises that “India needs a ‘clearer set’ of guidelines for approving FDI from China expeditiously as the system of case-by-case review is sluggish. Mirroring “small yard and high fence”, it is imperative that India examines FDI proposals from China through a security lens akin to the United States. Deep Seek is platformed on Nvidia H800 but is running inferences on the new Chinese chips made by Huawei (designated as a national security threat by the US) called the 910C. The 910Cs are an alternative to the H100.

In the upcoming national budget, the approach expected could be China collaborative, with greenlighting of investments but under stricter scrutiny to ensure local participation, encouraging local Indian enterprises. India could innovative and produce in the AI space, achieving advanced AI capabilities with cost-effective methods, notwithstanding systemic challenges within the typical complexities. But that’s the intention and energy behind Make in India and Design in India Design for the World. Deep Seek could inspire Indian innovators. Like Dep Seek, India could leverage its ability to refine and optimize open-source technologies. Innovations in AI does not always require the creation of new foundational models. India’s strength lies not in developing a match to ChatGPT or Deep Seek from scratch but in smartly building on what already exists. The mantra is to develop on global advancements while focusing on AI applications tailored to India’s typical needs.

Deep Seek is a Deep Ocean, hopefully dueling Deep Fake, in not necessarily a Deep State.

AI is the agent engulfing us, Deep Seek is its favourite sub-agent today.

(Charudutta Panigrahi is a thinker, author, and a technocrat. He can be contacted at [email protected])

 

DISCLAIMER: The views expressed in the article are solely those of the author and do not in any way represent the views of Sambad English.