China Probes Nvidia, Geopolitical Risks Weigh on Tech | Bloomberg Technology

China Investigates Nvidia, Impacting Market and US-China Tech Conflict

• China has opened an investigation into Nvidia, citing antimonopoly concerns over a 2020 deal in which Nvidia acquired an Israeli company that makes networking equipment, with the investigation potentially impacting Nvidia's market capitalization, which has already shaved off $100 billion, down 3% (1m4s).
• The deal in question was approved by China in 2020 with seven conditions, including giving customers in China access to information from Nvidia's own chips, and it is unclear what specific aspects of the deal China will investigate (2m0s).
• Nvidia has been at the center of the US-China conflict over technology, with Washington cutting off Nvidia's ability to sell its most advanced chips in China, and the Justice Department also looking into Nvidia for potentially violating laws (2m26s).
• The antimonopoly concerns surrounding Nvidia are not limited to China, as the US, France, and other countries are also looking into the company due to its near 100% market share in the most advanced chips used to train AI models (3m13s).

US-China Trade Conflict and its Impact on Companies

• The investigation is part of a broader trade conflict between the US and China, which has created problems for companies operating in both countries, including Nvidia, which sells a significant number of chips to China, but has been restricted by US regulations (3m25s).
• China has retaliated against US restrictions by cutting off exports of certain minerals, such as gallium and germanium, which are important to the chip industry and the defense industry (4m2s).

Trump and Shi Conversation Highlights US-China Uncertainty

• The investigation into Nvidia comes as the US and China are experiencing increased tensions, with President-elect Trump recently stating that he had an exchange with Chinese leader President Shi, marking the first clear indication of direct contact between the two since Trump's re-election (4m22s).
• The conversation between Trump and Shi has highlighted the uncertainty surrounding the US-China relationship, with Trump's campaign promises including imposing tariffs of up to 60% on Chinese goods and 10% tariffs on Chinese goods due to the flow of fentanyl from China into the US (5m13s).
• Beijing is looking to the change in administration as a potential opportunity, but is also preparing for possible retaliatory measures from the US, creating a complex and uncertain environment for companies operating in both countries (5m48s).
• The incoming US president's tendency to be more transactional may lead to a different approach in engaging with China, possibly including a more overt approach to restrictions on sensitive minerals (5m52s).
• The US has imposed restrictions on China, and in response, China has recently imposed restrictions on some sensitive and critical minerals, which may impact US businesses that rely on these minerals as key inputs to their industries (6m5s).
• Former US President Donald Trump has stated that he cannot guarantee anything, including the removal of tariffs on China, and has suggested that tariffs are a way to gain leverage in other policy areas, including immigration (6m38s).
• Trump has also claimed that there is no cost to consumers from tariffs, despite being pressed on the issue during an interview (7m32s).

US Economic and Market Conditions

• The impact of tariffs on businesses is a concern, as they would have to consider passing on increased costs to consumers or finding alternative suppliers (7m15s).
• The US has seen a sticky inflation print, which is sending tech stocks lower, and the stimulus potential from China is being closely watched (8m23s).

Nvidia's Valuation, Growth, and Market Position

• Nvidia is still garnering the majority of economics in this cycle, despite having slightly reduced exposure to China in recent years (9m11s).
• Nvidia's valuation, in excess of $3 trillion, is vindicated by the opportunity despite issues with China, and the company's strength in hardware and software is a key factor in its success (9m55s).
• The software aspect of Nvidia's business is underappreciated, but it provides a longer tail in terms of the ability to continue to sell through and pulls up the margin structure of the company over time (10m17s).
• Nvidia is being probed by China, and geopolitical risks are weighing on the tech industry, but there is still room for growth in the GPU space, with some peers showing more stickiness (10m43s).
• There has been a shift in chip stocks and semiconductors towards the software space, driven by the potential of general AI, but it's still early stages and the impacts are yet to be seen (10m55s).

AI, Advertising, and Investment Strategies

• Enterprise adoption of AI is slower, but the advertising space is seeing immediate returns, with the use of structured data to predict consumer behavior and match media content with advertising areas (11m31s).
• The focus has been on the portfolio's positioning in the advertising space, where real returns are being seen, and the economics are going in that direction (11m52s).
• A neutral posture has been taken, with some decent sizing in other areas of the economy, such as consumer staples and utilities, which offer more attractive multiples and returns (12m22s).
• A barbell approach is being used, with punchier names and higher conviction names, as well as good quality compounding at a reasonable price (13m4s).

Super Micro and Alphabet's Quantum Computing Advancements

• Super Micro shares jumped earlier, with Nasdaq granting the firm more time to become compliant with listing moves and financial filings (13m24s).
• Alphabet has revealed new details on the progress in quantum computing, including a new chip and potential advancements, with the biggest challenge being error correction (14m21s).
• A new high-quality quantum chip has been developed, enabling breakthroughs in computation and error correction, with the error rate coming down exponentially as more qubits are used (14m56s).
• A benchmark computation was performed, taking five minutes with the new chip, compared to 10 trillion years with a top supercomputer, using a benchmark random circuit to compare different processes (15m20s).
• The benchmark problem is not a real-world problem, but a suitable benchmark to compare different processes and chips against classical computers (15m54s).
• Quantum computing will be a transformative technology that can help solve complex problems in various fields, including healthcare, pharmaceuticals, and engineering, by enabling simulations and computations that are currently unsolvable with traditional computers (16m21s).

Quantum Computing's Potential and Impact

• Quantum computing is not a replacement for AI, but rather a complementary technology that can help AI work better, and vice versa, with many computations in AI where a quantum chip would be the tool of choice (17m57s).
• The application of quantum computing is not limited to healthcare, but can also be used to design fusion reactors, improve batteries for electric cars, and solve other complex engineering challenges (17m15s).
• The development of quantum computing is expected to have a significant impact on various industries, including energy and infrastructure, with the potential to create new challenges and opportunities (17m43s).

US Tech Leadership, AI Growth, and Infrastructure Challenges

• The US has a crypto-friendly president who will appoint an AI czar, representing the best technology for the future, and encouraging other countries to follow the US lead (20m16s).
• The growth of AI-powered products, such as those developed by Block Stone, has created difficulties for local communities, including the need for large amounts of power and the construction of transmission lines (20m52s).
• The construction of data centers, such as the one being built by QTS, requires significant amounts of power, with the potential to impact local communities and the environment (21m25s).
• The scaling of AI and data centers will require the construction of more power infrastructure, which may not be welcomed by local communities, but is necessary to support the growing demand for AI and data storage (21m35s).
• The development of AI and data centers is expected to continue, with no indication of slowing down, despite the challenges and difficulties associated with their growth (21m55s).
• The construction of data centers and transmission lines is driven by the need for faster and more efficient data processing, with power companies having the authority to build transmission lines to provide the necessary power (22m8s).
• The growth of AI and data centers has raised concerns about the impact on local communities, including the need for large amounts of power and the potential for environmental damage (21m49s).

Investigation into Migrant Labor and Tech Industry Concerns

• The investigation into migrant team labor at a facility in Illinois has raised concerns about the treatment of workers in the tech industry (19m46s).
• The allegations of migrant team labor at a facility in Illinois are being investigated by the Labor Department, with the potential for significant consequences for the company involved (19m58s).
• The development of AI and data centers has created new challenges and opportunities for companies, including the need to balance growth with social and environmental responsibility (21m8s).

Tech Industry Risk Aversion and China's Optimism

• The tech industry is experiencing a buildup in risk aversion, with a 0.7% decline, amidst concerns over inflation and Fed policy in the United States (23m10s).
• A new probe announced by China is creating anxiety for Nvidia, contributing to the risk aversion and a selloff in riskier assets, with the S&P 500 down 40% (23m28s).
• Despite this, there is optimism in China due to fiscal policies, with Morgan Stanley reporting the most positive outlook in over a decade, and Chinese names performing well in the market (23m47s).

Reddit's AI Chatbot and TikTok's Potential Ban

• Reddit is testing a new AI-powered chatbot called Reddit Answers to service discussions across the platform's forums, aiming to keep users within the platform (24m17s).
• The chatbot uses large language models from OpenAI and Google, making it easier for users to search and find information on the Reddit site (25m4s).
• Reddit has been signing deals with generative AI companies, including Google, to use their models and data, and is also building its own models (25m11s).
• TikTok's parent company, ByteDance, faces a ban in the US next month due to a federal appeals court decision, sparking opposition from civil liberties groups and users (26m7s).

US Government's Concerns and Negotiations with TikTok

• The ban is based on national security concerns, with the US government arguing that the Chinese government might manipulate the app to push propaganda or surveil Americans (26m58s).
• To address these concerns, the US government has been negotiating with TikTok to locate its data within Texas and implement measures to mitigate the risks (27m15s).
• The US government has expressed dissatisfaction with the level of protection over data, specifically regarding the Chinese government's potential manipulation of the TikTok app, despite having no concrete evidence of such manipulation (27m27s).

Legal Challenges and Precedents in TikTok Ban Case

• The BC Court of Appeals has ruled that the government's distrust of ByteDance and TikTok is enough to potentially ban the app from the country, a decision that may be posed to the US Supreme Court (28m7s).
• The case is unusual, with the closest precedent being a 1960s case involving Chinese communist propaganda, in which the Supreme Court ruled that Americans have the right to receive such propaganda if they choose to (28m58s).
• The Pentagon Papers case is another relevant example, where the Supreme Court ruled that the New York Times had the right to reveal the secret history of the Vietnam War, despite the government's claims of national security risks (29m11s).
• The argument that will be presented to the Supreme Court is that the government's claims of national security risks are not enough to justify banning the app, and that the court should not defer to the judgment of the political branches of Congress and the President (29m21s).
• The government has been trying to make its case against TikTok for the past six months, but Americans have continued to use the app despite the government's warnings, with many politicians, including Donald Trump, Joe Biden, and Kamala Harris, using the app themselves (30m38s).
• The government has stated that it has no evidence of foreign manipulation of the app, and it is unclear whether the government's concerns will be enough to justify a ban (31m5s).

TikTok's Impending Ban and Venture Market Bifurcation

• January 19 is the deadline for TikTok to be sold or banned, and it is increasingly likely that the app will be banned on that date (31m24s).
• The venture market is bifurcating into two different strategies, with multi-stage firms and mega funds on one side, and specialized, targeted strategies on the other, creating a barbell effect (33m9s).
• The representation of mega funds in the venture market has increased from 2% to 25%, leading to overrepresentation and implications for limited partners (LPs), managers, and founders (33m52s).
• This trend is expected to lead to more consolidation among early-stage funds and more specialization at the early stages, with firms needing to differentiate themselves to capture true alpha (34m20s).

Venture Capital Strategies and Mega Funds' Impact

• Multi-stage strategies often play an index game, putting out a lot of checks and backing winners, which can dilute the early-stage class and make it harder for smaller firms to compete (34m40s).
• To counteract this, firms like Underscore VC are focusing on early, non-consensus investments and digging deeper to find true alpha (34m57s).
• The impact of mega funds on early-stage bets can be multiple, including diluting returns across the board, but also creating opportunities for firms that can get in early or invest in non-consensus ideas (35m34s).

Founders' Strategies and AI Opportunities

• Founders are becoming more sophisticated in thinking about who they bring around the table at different stages of company building and considering questions of alignment and investment (35m53s).
• Most founders are having to lean into generative AI, even if they are not AI-specific companies, and this is creating opportunities for firms to uncover less covered areas with non-consensus builders and founders (36m36s).
• There is a ton of opportunity in AI, and firms are looking to capitalize on this trend (36m52s).

AI Investment Trends and Boston's Ecosystem

• Over 30% of venture capital in the past year went into AI companies, and this trend is expected to increase next year, with companies being built with new models and capital structures that allow for more efficient operations (36m53s).
• The mentality of doing more with less and achieving more with less is prevalent, and it's possible to build a big company with a different capital structure given the increased efficiencies of building with AI (37m21s).
• Many AI companies are being built across the US, but Boston is a particularly attractive ecosystem due to its production of top talent and its ability to solve hard problems in areas like healthcare, biotech, and financial services (37m47s).

Waveform AI: AI Audio Software and Business Model

• Waveform AI is a company that makes AI audio software with human-like voice features, and despite being small with only five people, it has a high evaluation and is working on software to improve human-computer interactions (38m42s).
• The company's founder, Alexis Kono, has a strong background in voice and computer interactions, having worked on OpenAI's advanced voice mode, and investors are betting on his expertise (39m11s).
• Waveform AI's business model involves releasing B2B and consumer-oriented products, potentially including an app for consumers to test the technology, and working with other companies to use their software (39m59s).

Oracle's Cloud Business and Earnings Outlook

• Oracle is set to release its earnings after the bell, with its stock having its best year since 1999 thanks to the momentum of its cloud business, and it remains to be seen if this momentum will continue (40m33s).
• Oracle's cloud business has done relatively well compared to competitors, and the company has become a serious player in the cloud space within the last two years (41m13s).
• Oracle's cloud has been growing rapidly due to its ability to handle demanding AI workloads, which traditional giants like AWS and Azure struggle with, making it the fourth hyper-scaler (41m27s).
• Oracle has some key clients, including Uber and TikTok, but concerns around the TikTok deal pose a risk, as a change in ownership could lead to a switch in cloud infrastructure (41m51s).
• However, the risk is not as acute as it once was, as Oracle now has a more diverse customer base (42m21s).
• Oracle has also been successful in managing its database segment, allowing customers to run its iconic database on Azure and AWS, which has breathed new growth into the segment (42m39s).
• The company has been building data centers across the world, but the amount of hype around its AI-centered rollout is massive, and the question remains how quickly it can convert demand into actual services (43m18s).
• Oracle faces challenges in meeting demand, including getting data centers up quickly and obtaining enough chips, similar to Microsoft and Amazon (43m44s).
• Despite these challenges, Oracle's shares have seen an 80% run-up, and the company's performance has been exciting for investors (43m14s).