Is ChatGPT Overhyped, or a New Era of AI?

By HYCM

What areas of the broader market are likely to be affected? What products are likely to emerge? Is it just the newest hype cycle similar to VR and crypto?

It is a good time to present some of the events since the ChatGPT story broke into the mainstream, as well as to test the technology, and to provide a market angle.

What Is ChatGPT?

In its own words, after being asked to describe itself:

“I am ChatGPT, a large language model trained by OpenAI. I am designed to answer a wide range of questions and engage in conversations on a variety of topics. I use natural language processing and machine learning algorithms to understand the context and meaning behind the words you input and provide relevant and accurate responses. While I am not a sentient being and do not have emotions or beliefs, I strive to provide helpful and informative responses to the best of my abilities.”

ChatGPT was developed by OpenAI. Formerly a non-profit organisation, it was founded in December 2015 with a joint pledge of $1 billion by a group including Ilya Sutskever, Elon Musk and Peter Thiel, as well as companies such as Amazon Web Services, Infosys, and YC Research. The initial idea behind the partnership was the development and promotion of benign AI that would prove to be beneficial to the world.

What Has Happened Since?

In 2019, OpenAI changed to a “capped for-profit” organisation. Microsoft made a $1 billion investment in the new OpenAI LP entity that same year. In December 2022, OpenAI launched a freely available version of ChatGPT, which saw a massive surge in usage as the public rushed to experiment with it.

In January of this year, Microsoft invested further in OpenAI, this time a rumoured $10 billion to be released over a multi-year period. The deal will see Microsoft eventually owning 49% of OpenAI, with the company running ChatGPT on Microsoft’s Azure platform.

Microsoft followed this up with the launch of “New Bing” in February, a version of their web search portal, which, according to the company, is: “a new, next-generation OpenAI large language model that is more powerful than ChatGPT and customized specifically for search.”

In the meantime, recognising the potential hit to its own established search business, Google rushed to bring Bard, its own language-based AI, into the public awareness with a poorly prepared and badly received demo in which the AI made mistakes. The demo is said to have led to a $100 billion drop in Alphabet’s market cap.

Microsoft’s own New Bing launch didn’t fare much better, with criticisms emerging that not only did its algorithm provide inaccurate responses, but that it was also argumentative, and even rude.

It’s clear that we’re witnessing the very earliest stages of what could be a transformative technology, which nevertheless has a long way to go. It was to be expected that search would be one of the first for-profit deployments, especially considering how lucrative a vertical it is. Last year Google generated $162 billion from its search business. Whether search proves to be AI’s “killer app” remains to be seen.

How Accurate Is ChatGPT?

This is a key question which will affect how popular and effective such tools become. For example, during Google’s own Bard AI demo, when asked “what new discoveries from the James Webb Space Telescope can I tell my 9 year old about?” it included the following inaccuracy among a number of other bullet point responses: “JWST took the very first pictures of a planet outside of our own solar system.” This kind of demo spooks investors. Is AI really an information edge? Or will inaccuracies plague it?

When we asked ChatGPT: “When will Jerome Powell next cut the US Fed funds rate?” It initially stated that as an AI language model it possesses no insider information or ability to predict FOMC decisions. It went on to synopsise the basic factors influencing interest rate decisions, but it then made the following error:

“That being said, I can tell you that the Federal Reserve has indicated that it plans to maintain a loose monetary policy for the time being, with interest rates likely to remain low to support economic growth and recovery. However, any changes to interest rates will ultimately depend on the economic conditions at the time of the FOMC meetings and the decision of the committee members.”

ChatGPT failed to identify the current tightening cycle. So, there are clearly some limitations to the technology at this point in time.

Google Vs Microsoft

Broadly, this story appears to be an AI arms race between Microsoft and Google. Microsoft’s investment in OpenAI puts it ahead of Google in terms of AI chatbot technology, but Google is way ahead in terms of its search and advertising businesses. The former does, however, pose a threat to the latter should Microsoft succeed in creating a product that possesses better search capabilities than Google Search, which has almost become the western world’s de facto entrance point to the Internet.

Microsoft also has a far more diversified revenue stream than Google, with only 6% of its revenues coming from search, to Alphabet’s almost 60%. With that in mind, this specific struggle is far more existential to Alphabet than it is to Microsoft, which may go some way to explaining Google’s haste in debuting its own rival to ChatGPT.

Currently these demonstrations don’t amount to much more than branding exercises, but as the technology reaches mass adoption it could lead to changes in fortunes among the big tech companies similar to how Google outcompeted Yahoo in the early days of the Internet.

Technically, Microsoft and Google’s recent price action can be said to reflect the above story. Alphabet is currently trading below all major long-term moving averages as it re-tests multi-year lows for the third time since 2022. Meanwhile Microsoft, while still clearly in a downtrend with the rest of the broader market, appears to have found support at its own 200-day moving average, and is in the process of setting a lower-high, testing its 20-week moving average as support.

Nvidia

As some of the early 2023 hype surrounding the technology begins to fade, investors are looking at some of the “pick-and-shovel” investments related to artificial intelligence. Pick-and-shovel investing refers to a focus on the companies that provide equipment to an industry, rather than the industry itself. In the case of a gold rush, it would be actual picks and shovels. In the case of AI, it’s the semiconductors that are crucial to providing these services at scale.

Rather than guessing on which company will harness AI to the most profitable effect, investors appear to be searching for companies that are set to outperform regardless of the above struggles.

Nvidia has been one of the biggest benefactors of the surging interest in AI spurred by the ChatGPT release. The chips it produces are not only necessary for AI applications, but also crucial to data centres, so the stock is a bet on both AI and cloud infrastructure. This comes as major crypto projects like Ethereum move away from proof-of-work, and thus no-longer require GPUs for mining. Nvidia recently provided a modest beat to earnings expectations ($6.05 billion) and expects Q1 revenue to grow to $6.5 billion.

The stock is one of the few tech charts that seem to suggest that the bear market is over. It has significantly outperformed the benchmark, having cleared all its major long-term moving averages, found support at the 20-week MA and set a higher-low there. It has also rallied over 100% from the bottom and set an impressive higher-high that’s within 50% of the stock’s all-time high.

Testing ChatGPT

When asked about the trends in the financial markets, the chat bot listed recovery from the pandemic, inflation concerns, increased focus on sustainability, digital transformation, and volatility in cryptocurrencies.

It also provided several changes the retail forex industry has undergone over the last few years. Here are some of the notable changes: increased regulation, significant advancements in technology, which have made trading more accessible, efficient, and user-friendly, increased competition which led to lower trading costs, better services, and more options for traders, the emergence of social and cryptocurrency trading, the use of artificial intelligence and machine learning leading to the development of sophisticated trading algorithms that can help traders make more informed trading decisions.

When asked about HYCM, a worldwide brand under the HYCM Group, ChatGPT provided the following description. It has been cut down for the sake of brevity, without being edited:

“HYCM is a global online trading platform that offers trading services in a variety of financial instruments, including forex, stocks, indices, commodities, and cryptocurrencies*. The company was founded in 1977 and is headquartered in the United Kingdom, with additional offices in Cyprus, Dubai, Hong Kong, and Kuwait.

HYCM is regulated by several reputable financial regulatory authorities, including the FCA, CySEC and DFSA. This regulation ensures that HYCM adheres to strict standards of financial transparency and security, helping to protect the interests of its clients.

Overall, HYCM is a well-regulated and reputable online trading platform that provides traders with a wide range of trading instruments, advanced tools, and excellent customer support.”

Future Outlook

There is a growing demand for AI-powered solutions in various industries with an aim to help companies improve efficiency, productivity, and customer service. However, there are also concerns about the potential risks associated with AI technologies, including bias, privacy, and security issues.

Overall, the future of models like ChatGPT is promising, but it is essential to continue to develop and refine these models to ensure that they are ethical, transparent, and secure.

Trade with HYCM

*Note: Cryptocurrencies are not available for trading under HYCM (Europe) Ltd and HYCM Capital Markets (UK) Limited.

About: HYCM is the global brand name of HYCM Capital Markets (UK) Limited, HYCM (Europe) Ltd, HYCM Capital Markets (DIFC) Ltd, HYCM Ltd, and HYCM Limited, all individual entities under HYCM Capital Markets Group, a global corporation operating in Asia, Europe, and the Middle East.

High-Risk Investment Warning: Contracts for Difference (‘CFDs’) are complex financial products that are traded on margin. Trading CFDs carries a high degree of risk. It is possible to lose all your capital. These products may not be suitable for everyone and you should ensure that you understand the risks involved. Seek independent expert advice if necessary and speculate only with funds that you can afford to lose. Please think carefully whether such trading suits you, taking into consideration all the relevant circumstances as well as your personal resources. We do not recommend clients posting their entire account balance to meet margin requirements. Clients can minimise their level of exposure by requesting a change in leverage limit. For more information please refer to HYCM’s Risk Disclosure.

*Any opinions made in this material are personal to the author and do not reflect the opinion of HYCM. This material is considered a marketing communication and should not be construed as containing investment advice or an investment recommendation, or an offer of or solicitation for any transactions in financial instruments. Past performance is not a guarantee of or prediction of future performance. HYCM does not take into account your personal investment objectives or financial situation. HYCM makes no representation and assumes no liability as to the accuracy or completeness of the information provided, nor any loss arising from any investment based on a recommendation, forecast, or other information supplied by an employee of HYCM, a third party, or otherwise.