Using AI to generate stock pitches and other finance stuff
I have been playing with the chatGPT feature in open.ai, and It is somewhat cool to see an AI generate a stock pitch. Its very rudimentary but I could see this being useful especially if you want to get a brief overview of what a company does for your own reading.
Some examples I generated:
Prompt: Give a stock pitch on Coca Cola
OpenAI Response:
Coca-Cola is a global beverage company that produces and sells a wide range of non-alcoholic beverages. The company is best known for its flagship product, Coca-Cola, a carbonated soft drink that is one of the world's most recognizable and valuable brands. Coca-Cola is the world's largest beverage company and is a member of the Dow Jones Industrial Average. The company operates in more than 200 countries and has a diverse product portfolio that includes more than 500 brands, including sparkling beverages, juice, water, sports drinks, and coffee.
One of the key strengths of Coca-Cola is its strong brand recognition and customer loyalty. The company has a long history of successful marketing campaigns that have helped to build its brand and create a loyal customer base. Coca-Cola has also made significant investments in its production and distribution network, which has helped it to maintain a strong presence in global markets.
Another strength of Coca-Cola is its diversified product portfolio. The company offers a wide range of non-alcoholic beverages, which allows it to appeal to a wide range of consumers and to respond to changing consumer preferences. This diversification also helps to reduce the company's dependence on any one product or market, which can be beneficial in times of economic uncertainty.
Despite these strengths, Coca-Cola does face some challenges. The global beverage industry is highly competitive, and the company faces strong competition from other beverage companies. In addition, health concerns about sugary drinks have led to increasing scrutiny of the industry, and some consumers are shifting to healthier beverage options.
Overall, Coca-Cola is a well-established and highly recognizable brand with a diversified product portfolio and a strong global presence. While the company does face some challenges, it is well-positioned to continue to grow and generate value for its shareholders.
Prompt: Write an equity research report on Apple
OpenAI Response:
Apple Inc. is a leading global technology company that designs, develops, and sells consumer electronics, computer software, and online services. The company's best-known products include the iPhone, iPad, Mac, and Apple Watch, as well as the iTunes, iCloud, and Apple Music services.
Apple has a strong financial track record, with a history of steady revenue and profit growth. In its most recent fiscal year, the company reported revenue of $274.5 billion, an increase of 21% compared to the previous year. Net income for the year came in at $57.4 billion, an increase of 31% compared to the previous year.
Apple's strong financial performance has been driven by the continued success of its flagship iPhone products, as well as the growing popularity of its services segment, which includes the App Store, iCloud, and Apple Music. The company has also been expanding into new markets, such as wearables and home automation, through products like the Apple Watch and HomePod.
Looking ahead, Apple is well-positioned to continue its strong performance. The company has a loyal customer base and a strong brand, and its diverse range of products and services provides a steady stream of revenue. Additionally, the growing demand for smartphones, tablets, and other connected devices is expected to continue to drive growth for the company.
Overall, Apple is a solid investment opportunity. The company has a strong financial track record and a diversified range of products and services, and is well-positioned to capitalize on the growth of the global technology market. A reasonable price target for Apple's stock could be around $150 per share
Prompt: Walk me through a DCF
OpenAI Response:
A discounted cash flow (DCF) analysis is a method used to estimate the intrinsic value of an investment by calculating its future cash flows and discounting them back to the present value. This method is based on the idea that the value of an investment is the sum of all of its future cash flows, discounted at a rate that reflects the time value of money and the level of risk associated with the investment.
To perform a DCF analysis, the first step is to forecast the investment's future cash flows. This typically involves making assumptions about the growth rate of the investment's revenues and expenses over time, as well as the timing of those cash flows. Once the future cash flows have been forecasted, the next step is to determine the appropriate discount rate to use in the analysis. This rate should reflect the time value of money and the level of risk associated with the investment.
Once the future cash flows and discount rate have been determined, the next step is to discount the future cash flows back to their present value. This can be done using a simple present value calculation, which takes the future cash flow and divides it by the discount rate plus one, raised to the power of the number of periods until the cash flow is received. For example, if a future cash flow of $100 is expected in three years and the discount rate is 10%, the present value of that cash flow would be $100 / (1 + 0.1)^3 = $75.36.
Once all of the future cash flows have been discounted to their present value, the final step is to sum all of the present values to arrive at the investment's intrinsic value. This value can then be compared to the investment's current market price to determine whether the investment is undervalued, overvalued, or fairly valued.
Overall, a DCF analysis is a useful tool for estimating the intrinsic value of an investment and determining whether it is a good value at its current market price. However, it is important to remember that the accuracy of the analysis depends heavily on the assumptions used in forecasting the investment's future cash flows and selecting the appropriate discount rate.
Prompt: Why and when should I invest in consumer cyclicals
Investing in consumer cyclical companies can be a good way to capitalize on the growth of the economy and consumer spending. These companies are sensitive to changes in the business cycle, and tend to perform well when the economy is strong and unemployment is low. This can lead to increased revenue and profit growth for these companies, which can translate into strong returns for investors.
However, it is important to keep in mind that consumer cyclical companies are also subject to the risks of an economic downturn. When the economy slows and unemployment rises, consumers tend to reduce their spending, which can hurt the performance of these companies. Therefore, it is important for investors to carefully consider the state of the economy and the potential risks before deciding to invest in consumer cyclicals.
Overall, investing in consumer cyclicals can be a good way to capitalize on the growth of the economy and consumer spending. However, it is important to carefully consider the risks and be prepared for the potential impact of an economic downturn.
This stuff is pretty nifty, but I am wondering how many people in the future are just gonna give a stock pitch generated from an AI during an interview. I have heard accounts on Twitter of people in school submitting papers generated with AI and they pass with A's, undetected by plagiarism scanners.
I wonder what this could mean for investing/research roles in the future when this tech is more mature.
Don’t let banks see this. They are gonna layoff more analysts
Bump. Interesting
Super interesting! Didn’t know this was possible
ChatGPT, which just came out, is a major leap forward in usability and performance. This is a pretty big deal.
Some guy just fed it an entire practice SAT exam. It got a 1020. Engineering the prompts better would probably improve the score. And future models will be better.
https://twitter.com/davidtsong/status/1598767389390573569
Homework, in the traditional sense, is about to become obsolete. Along with a lot of other things.
Wut? homework isn't becoming obsolete
Just create an account on OpenAI's site and start feeding some typical school essay prompts and homework questions into the chat interface. It's already, right now, better than a huge number of students. And it churns out answers in seconds.
Wow - this is cool. Much better than I expected.
I really hope someone out there ends up building some software that replaces 80% of analysts and associates. If AI can paint portraits, then it can definitely crunch numbers, build scenarios, and digest 10-K. You only put 2 analysts (instead of 10) who would be responsible to verify that the model makes sense and adjusting it doesn't.
If someone ends up building this, he could easily ask $5M - $10M annual fee subscription per bank considering that it replaces 80% of analysts and associates (around 500 people who get paid in total ~ $140M per year).
Even if the heavy work is done by AI, high fees should remain, the justification being "It's not about what AI builds, it's about interpreting those numbers".
A true paradise in terms of income for MDs and VPs.
Curious how did you get it to run? When I try to enter the same prompt I get "I'm sorry, but I am not able to provide you with information about specific companies or their stocks. I am a large language model trained by OpenAI and do not have access to real-time information about companies or their stocks. I am only able to provide general information and answer questions to the best of my knowledge based on the information available to me."
yeah, I started having that problem too. Earlier this week I was able to ask it to pitch me stocks in specific categories (mid-cap biotech, etc) and it would give me a specific company. Now it just says ABC Company, XYZ company, not any real company name. They must have changed it to stop spitting out stock pitches. Sometimes you have to trick it into saying what you want it to say
Below applies to buyside / sell-side equity research (not IB):
No offense but this is completely boiler-plate and doesn't tell you anything that just googling a few articles would surface. Even crappy sell-side is far better (in terms of what investors will use to get up to speed on a company), much less the more differentiated analysis the better analysts can produce...and this is ultimately inferior still to good buyside research which necessitates digging even deeper
Maybe this is just the first step and in 10yrs this tech gets to that point, but until this AI can at least be better than a shitty SS analyst I doubt it's getting much adoption
i mean what were you expecting? a generalist chat-based language model that just so happens to be able to perform highly specialized job like ER at the level of a human? its confusing because you say "no offense but..." even though quite literally noone on this thread has suggested that this would replace an analyst.
The Apple numbers are completely wrong. I think this AI is helpful as a starting point though as a template.
One could also just use the past company’s report.
Yea the model can only go up to 2021 right now and it can’t search the internet either I don’t think
Only bankers would think this is industry-changing… aka the sellside monkeys who are trained to think in surface level, Type I thinking that any talented high schooler can do. Any competent buyside equity research analyst would laugh at this shit..
chatGPT is very impressive. Who knows where this technology will be 10 years from now?
Just used open ai to pitch me a merger between Wendy’s and five guys for fun and I was so surprised with the result:
“This merger would provide a delicious combination of fast food and burgers that customers love. Wendy's is known for its classic burgers with square patties, while Five Guys is known for its made-to-order burgers that are customizable with an array of toppings. By merging, customers could enjoy the best of both worlds in one convenient location. The combination of Wendy's and Five Guys would expand the product offerings, allowing customers to choose from an even wider selection of burgers and other fast food favorites. Additionally, the merger would create cost savings and efficiencies, resulting in lower prices for customers. With this merger, Wendy's and Five Guys can become the leading provider of fast food and burgers, providing customers with the best of both worlds.”
This sounds exactly what it would be like on businesswire or the execsum. Will not be surprised if a lot of jobs will be replaced by AI. It is scary.
Another one
Prompt:
Write an equity research initiation report on Apple. This report should be long, and the reader should be able to understand everything that apple does. This report should match the quality of top sell side research analysts.
OpenAI response:
It can even draft real estate leases
Prompt:
Draft a commercial lease in NJ for a property that is selling for 150 per square ft
OpenAI response:
I think this is honestly really cool. ChatGPT is very surface-level, boilerplate and basic, and it even makes blatant errors here and there - but personally, i think the biggest takeaway for equity research is that it tells us what 'the Street' is thinking. All that surface level shit, it's the consensus thought. We can easily use it as a proxy for consensus, know where the market is at, and know where our view stands in relation to consensus.
Of course, this is an extremely crude measure and I'm just throwing my thoughts out there, but it's a good starter. We all know 'consensus' is everything in the public markets. ChatGPT would be '1st level thinking' as Howard Marks would put it - it tells analysts what the 1st-level thinkers are thinking, and allows us to focus on being 2nd or even 3rd level thinkers
Incidunt id nihil quas. Illo laborum consequatur earum sed sit error eaque. Adipisci libero velit vel aliquam facilis omnis. Voluptas recusandae officia aut distinctio. Quas praesentium dolor saepe adipisci et alias.
See All Comments - 100% Free
WSO depends on everyone being able to pitch in when they know something. Unlock with your email and get bonus: 6 financial modeling lessons free ($199 value)
or Unlock with your social account...