• AI Weekly
  • Posts
  • How to Be A Better Investor Using A.I

How to Be A Better Investor Using A.I

Full guide inside.

In partnership with

Get The Crypto Playbook for 2025

Keeping up with crypto while working a full-time job? Nearly impossible.

But Crypto is on fire and it’s not slowing down, with the industry having just hit a record-high $4 trillion dollar market cap.

And we’re sharing it at no cost when you subscribe to our free daily investment newsletter.

It covers the new Crypto bills that just passed and all the top trends, including the altcoin we think could define this cycle. That’s right, you can catch up on the industry in 5 minutes and still take advantage of this record bull run.

Skip the noise and stay one step ahead of the crypto and stock market.

Stocks & Income is for informational purposes only and is not intended to be used as investment advice. Do your own research.

How to Use AI as a Better Investor (and Invest in the AI Revolution)

AI is taking the world by storm, and if you’re into investing, you’ve probably wondered two things: (1) Should I invest in AI companies for the long run? and (2) Can I use AI tools (like ChatGPT) to make smarter investing decisions? This guide will walk you through both. We’ll keep it conversational and easy to follow – no PhD in computer science required. Let’s dive in!

Why All the Buzz About AI in Investing?

Imagine living through the late 1990s dot-com boom. Lots of hype, huge bubble, big bust... yet eventually the internet gave rise to giants like Amazon, Google, and Facebook. AI (Artificial Intelligence) today is often compared to that era. Everyone’s talking about how AI could transform industries from healthcare to finance to (yes) personal gadgets. Some redditors even joke that all this progress is really just leading to “lonely people finally getting robot companions” (you can guess what kindreddit.com). Jokes aside, AI is a big deal because:

  • Widespread Impact: AI isn’t one thing – it’s a wave of tech changing how we drive cars, diagnose diseases, recommend products, and more. That means opportunities in many sectors.

  • Productivity Boost: AI can crunch numbers and find patterns faster than any human. Companies adopting AI might become more efficient and profitable.

  • Hype vs Reality: Like the internet boom, not every AI idea will succeed. For every Amazon, there were dozens of Pets.com that went bust. The challenge (and opportunity) is figuring out who the future winners might be.

In short, investors are excited about AI because it could be as revolutionary as past tech booms – and those who invest wisely might see huge returns over 10–20 years. But how do you do that without betting on the wrong horse? Let’s learn from the past.

Investing in AI for the Long Term: Lessons from the Dot-Com Era

If you had $1,000 back in 2000, would you have known to put it in Amazon or Apple? Hindsight is 20/20, but at the time it was super hard to predict which dot-com companies would survive and thrive. The same goes for AI today: we know it’s a game-changer, but we can’t be sure which specific companies will dominate a decade from now. Here are some tips to increase your odds:

  • Diversify Your Bets: One Reddit user put it bluntly: “For every successful FAANG, there are 1000s of failed attempts.”sciencedirect.com Instead of trying to pick one magic AI stock, consider a basket of companies or an ETF (exchange-traded fund) focused on AI or tech. For example, an ETF like Nasdaq-100 (QQQ) or a tech fund (such as Vanguard’s VGT) gives you exposure to many tech firms at once. There are even AI-themed ETFs (some focus on semiconductor chips, others on robotics or cloud computing).

  • Focus on the “Picks and Shovels”: During a gold rush, the guys selling shovels often made steady money. In the AI boom, the “shovels” are things like microchips, cloud services, and data centers. AI needs hardware and infrastructure. Companies like NVIDIA (NVDA), which makes the GPUs that AI models run on, or Taiwan Semiconductor (TSM) that manufactures advanced chips, have been hot investments. (In fact, NVIDIA’s stock skyrocketed as AI took off – it’s been a star of this trend.) Other examples include ASML (dominates high-end chipmaking equipment) and AMD or Intel (design chips too).

  • Data is Gold: Some say “data is the new oil.” AI learns from data, so companies with huge datasets and user bases could have an edge. Think of Google (and its parent Alphabet) – they’ve been investing in AI for years (Google’s DeepMind unit and AI research is top-tier), and they have tons of data from search, YouTube, etc. Amazon has consumer purchasing data and cloud computing (AWS). Meta (Facebook) has social data (though some feel its data might be less useful than Google’s for certain AI tasks). Apple and Microsoft are also in the game – Apple uses AI in devices, Microsoft invested heavily in OpenAI (the folks behind ChatGPT) and is weaving AI into its products. Chinese tech giants like Baidu (often called “China’s Google”) and Alibaba/Tencent are also big in AI. The idea is that companies that control a lot of data or provide AI services might prosper.

  • AI in Every Industry: Besides the obvious Big Tech names, look at industries being transformed by AI. For example, healthcare – AI is helping discover new drugs and diagnose diseases. Companies focusing on AI drug discovery or medical imaging AI could boom. Automotive – self-driving car tech (Tesla, Waymo under Alphabet, or smaller players). Manufacturing and Robotics – factories are using AI-powered robots (companies like ABB or Fanuc, or newer ones making humanoid robots). Finance – fintech firms using AI for automation or trading. Even agriculture – AI-driven drones and farm equipment! You might consider a few leaders (or an ETF) in these niches if you believe in those applications.

  • Beware the Hype Cycle: Remember how the internet was hyped in 1999, crashed in 2000-2002, but truly changed life by 2010? AI might similarly have ups and downs. There could be bubbles (some AI startups or stocks are arguably overpriced now just due to excitement). As one commenter noted, “We all know AI will be ‘disruptive’, but where? How? Few are smart enough to predict the things AI would unlock.”sciencedirect.com So keep a long-term perspective. Don’t go “all-in” on one risky AI stock thinking it’s a guaranteed lottery ticket. It might fail or take longer than expected to succeed. Instead, pace yourself and consider dollar-cost averaging into solid investments.

Pro tip: If you’re unsure which company will win, one safe approach is broad index funds. An S&P 500 index fund (like VOO) includes the biggest companies – many of which are adopting AI. One Redditor advised this: if AI truly transforms the economy, the whole S&P 500 will benefit; if it’s overhyped, at least you didn’t bet on a single bubble stock and your index will still hold valuewhitecoatinvestor.comyoutube.com. In other words, you “own the haystack” instead of picking one needle.

AI Stocks and Sectors People Are Talking About

To make it concrete, here are a few types of companies folks often mention for a 10–20 year AI horizon:

  • Chipmakers: NVIDIA, AMD, TSM, ASML (critical for all AI development). Also Broadcom (AVGO) supplies parts for AI hardware.

  • Tech Giants: Alphabet (GOOGL), Microsoft (MSFT), Amazon (AMZN), Apple (AAPL), Meta (META). They have AI research arms and integrate AI into products. They also have the deep pockets to acquire promising AI startups.

  • Enterprise AI & Software: Examples like Palantir (PLTR) (focuses on AI for data analytics, defense, etc.), Snowflake (SNOW) or Databricks (data platforms enabling AI work) – helping other companies use AI. Even legacy firms like IBM (with Watson) or Oracle are trying to stay in the game.

  • Autonomous and Robotics: Tesla (TSLA) for self-driving and robotics (they’re building a humanoid robot), Google’s Waymo (if Alphabet breaks it out), Uber (investing in AI for logistics), Boston Dynamics (not public, but robotics companies could emerge), or industrial robot makers.

  • Healthcare AI: e.g. companies using AI for drug discovery (Schrödinger, Recursion), medical device companies adding AI, or even big pharmaceuticals partnering with AI labs.

  • Cybersecurity: With AI, unfortunately, comes smarter cyber threats. Security firms like Palo Alto Networks (PANW), CrowdStrike (CRWD), or Fortinet (FTNT) are using AI to detect threats. (Though, note: not everyone is bullish on every name – some experts on Reddit call certain firms “memes” despite the buzz. Always do your homework.)

  • Others: Quantum computing (still early but potentially revolutionary), cloud software providers (AI-as-a-service platforms), and even gaming or social platforms (since they can leverage AI for content and engagement).

Keep in mind, the stock market is forward-looking. A lot of AI optimism might already be baked into current prices. For example, by late 2023, NVIDIA had a very high valuation because everyone knew it’s central to AI. That doesn’t mean it won’t keep growing, but expectations are high. Balance your excitement with valuation sense – sometimes great companies become bad investments if bought at crazy prices.

Using AI Tools to Invest Smarter: Your Personal Assistant for Finance

Now, onto the second big topic: How can you use AI (like ChatGPT and friends) to be a better investor? Think of AI as your research assistant. It’s like having a really fast, well-read (but sometimes overly confident) intern on call 24/7. Here’s how investors are tapping these tools:

1. Quick Explanations of Complex Concepts

Ever scratched your head over a term like “DCF model” or “Sharpe ratio”? AI chatbots can break down complicated financial concepts into simple terms. For example, you can ask, “What does free cash flow mean?” or “How do stock buybacks work?” and get an easy-to-digest answer. This is super helpful for learning the basics or refreshing your memory. One user said they use AI as a way to get a “quick refresher” on things they learned before but forgot – way better than bugging a colleaguereddit.comreddit.com.

2. Summarizing Mountains of Information

Investing means reading a lot: company earnings reports, financial statements, news articles, you name it. AI can help summarize these for you. For example:

  • Earnings Calls/10-Ks: You could copy-paste sections of a company’s quarterly earnings transcript or annual report into ChatGPT and say, “Summarize the key points. What risks did the company mention? How were the earnings vs last year?” In seconds, you get a digest. As one person shared, they use ChatGPT for “earnings report summaries, stock comparisons, and sector analysis” and find it very beneficialreddit.com.

  • News Summaries: Let’s say there’s breaking news about a company you follow. You can feed the article to an AI (some tools let you share a link or text) and ask, “What happened and does it seem positive or negative?” The AI might respond with a concise summary. In fact, studies have shown ChatGPT can analyze news headlines and gauge if they’re good or bad for a stock – sometimes its news-based predictions correlated with actual stock moves the next daybusinessinsider.combusinessinsider.com! (Impressive, but don’t get too excited – more on this later.)

  • Comparing Companies: You’re interested in, say, electric vehicle companies. Ask something like, “Compare Tesla, Ford, and NIO in terms of their electric vehicle strategy and financial health.” A good AI might produce a neat comparison: e.g. Tesla – profitable, growth in self-driving tech; Ford – transitioning from traditional cars, investing in EV, strong legacy revenue; NIO – newer company, rapid growth in China, not profitable yet. It’s like a quick analyst report synopsis.

3. Generating Investment Ideas and Scenarios

Sometimes you need inspiration or a second opinion. AI tools can brainstorm with you (in a non-biased way):

  • Idea Generation: “What are some emerging technologies in AI I should research for investment?” The AI could list things like AI in education, AI for climate tech, etc., which might point you toward companies or ETFs in those areas.

  • Portfolio Analysis: You can input your portfolio (e.g., list your holdings and percentages) and ask, “What does my portfolio skewer towards? Is it well diversified? What risks should I be aware of?” One person did this and the AI pointed out, for example, if they were too concentrated in tech or lacked international exposure. It’s like getting a second set of eyes on your allocation.

  • What-If Scenarios: “If I invest $1000 per month for 10 years with a 7% annual return, how much could I have at the end?” An AI can do the math and tell you roughly (that would come out to around $173k in that scenario). Or more complex: “If interest rates rise, how might that affect tech stock valuations?” The AI can outline the cause-effect (higher rates make borrowing costlier, possibly slowing growth, etc.). It’s not making a prediction per se, but it helps you think through scenarios.

4. Automating Repetitive Tasks

If you’re into coding or spreadsheets, AI can even help you write simple code or formulas to analyze data:

  • Writing a Screener: For example, “AI, help me write a Python script to pull historical stock prices for AAPL and calculate the 50-day moving average.” ChatGPT can often produce a workable script (using Python libraries) that you can run. Many users have done this to backtest strategies or crunch numbers without doing all the programming from scratch.

  • Excel/Google Sheets Formulas: “How do I write a formula to calculate the CAGR (compound annual growth rate) given these cells?” The AI can spit out the formula or even give step-by-step. It’s like having an on-demand tutor for finance math.

  • Alert Systems: Some advanced users connect AI to their data. One person mentioned using an AI assistant tied to their watchlist, which would notify why a stock in their portfolio moved (e.g. “Stock X is down 5% today after missing earnings estimates”) – essentially summarizing news for them in real-time. Tools like that (some mentioned an app called Alpha for investors) act like a personalized news analyst.

5. Education and Coaching

Think of an AI chatbot as a non-judgmental teacher. If you’re new to investing, you can have a full conversation: “Explain the difference between a Roth IRA and a 401(k) in simple terms.” You can keep asking follow-ups. It won’t get annoyed – it’s literally what it’s made for! Some people even use it to simulate Q&A about their strategies, like “Here’s my reasoning for buying XYZ stock. Play devil’s advocate and tell me what I might be missing.” A good AI might actually list some risks or counterpoints (e.g., “XYZ’s debt is high, which could be a problem if interest rates rise.”). This can be invaluable for spotting blind spots in your thinking.

Not all AI assistants are created equal. Here are a few you might come across, and what they’re best at:

  • 🔹 ChatGPT (by OpenAI): The one that started this hype. Great at understanding questions and giving detailed answers. The free version has knowledge up to late 2021, but newer ChatGPT-4 (the paid version) can use plugins or browsing (meaning it can access more recent info if enabled). People use ChatGPT to summarize filings, explain concepts, and even analyze data (with the Code Interpreter or Advanced Data Analysis feature). It’s like an all-purpose tool. Tip: Feed it specific info (like portions of a financial report) for better results, rather than asking super broad things like “what stock should I buy?” – that usually gives generic (and not very useful) answers.

  • 🔹 Bing AI (Microsoft Bing Chat): Powered by a version of GPT-4, but it’s connected to the live web (and it’s free). You can ask, “What’s the latest news on Apple stock?” and it will actually search the web and give a summarized answer with citations. This is handy for real-time info. It tends to be a bit more factual (since it cites sources) but might have shorter responses.

  • 🔹 Google Bard (now powered by Gemini model): Google’s answer to ChatGPT. It can also access up-to-date info and is improving quickly. Some users find it good for certain tasks like coding or getting current data.

  • 🔹 Perplexity AI: A lesser-known but cool tool. It’s like a supercharged search engine with a chatbot. You ask a question, it gives an answer with footnote citations you can click (so you can verify the info). For example, ask “What is Tesla’s current debt level?” – it will search and show an answer like “Tesla’s debt is $X as of 2023【source】”. People like Perplexity because it reduces the chance of just making stuff up – you can trace every fact to a sourcereddit.comreddit.com. The free version is solid; there’s a paid version with more advanced capabilities too.

  • 🔹 FinChat / BloombergGPT / Finance-Specific Bots: There are AI tools specifically trained on financial data. FinChat.io lets you chat with a bot about specific stocks, pulling from SEC filings, press releases, etc. It’s geared to answer things like “What were Microsoft’s earnings in Q2?” or “Summarize the risks from Amazon’s annual report.” These often have a cost, but they can be more accurate on finance facts because they’re specialized. (Bloomberg, the big finance news company, even built its own huge finance-focused AI model – not public, but it shows the trend.)

  • 🔹 Stock Ranking AI Platforms: Some services like Danelfin use AI to rank stocks by potential, giving them scores (e.g. “8 out of 10” for likelihood to outperform in the next quarter, based on thousands of indicators). These can be interesting to explore for idea generation, though take the “scores” with a grain of salt and always understand why a stock is ranked so – don’t just buy because an algorithm said “9/10”!

  • 🔹 AI in Brokerage Apps: Big brokers are integrating AI too. For instance, Schwab’s premium clients have an “AI-powered” tool called Schwab Genius (or something along those lines) that analyzes portfolios. Morgan Stanley built an AI assistant for their advisors. Even retail apps might add AI chat features to answer your questions about your account or a stock. Keep an eye out in the apps you already use – you might see a friendly chatbot popping up offering help with your investing questions.

  • 🔹 Custom Setups: Some tech-savvy investors create their own little AI workflows. For example, using Notebook LM (Google’s AI tool for notes) to feed in a bunch of research PDFs and ask questions across them, or using the Python API for OpenAI to process financial data in bulk. If you’re into programming, the sky’s the limit on customizing AI for your needs. But if not, no worries – there are plenty of ready-made tools as mentioned above.

A word of caution: Many new AI tools will appear claiming to predict stock prices with AI or do all the work for you. Be skeptical of bold claims. Remember that one Redditor humorously said: “Please, keep using ChatGPT to do your investment research. I need willing losers to trade with.”reddit.com The point is, if it were that easy to have an AI consistently beat the market, everyone would use it and then nobody would have an edge. Which leads us to an important discussion…

Pros and Cons of Using AI for Investing

Like any tool, AI has its advantages and pitfalls. Let’s break them down:

Pros (How AI Can Help You):

  • Saves Time: Instead of reading a 100-page annual report for an hour, you can get the gist in minutes. AI can crunch through large data or text quickly and spit out highlights.

  • Educational: AI can teach and clarify. It’s great for beginners to ask basic questions without feeling embarrassed, and great for experienced folks to explore ideas or get quick calculations.

  • Comprehensive Research: It can gather perspectives from all over. For example, if you ask, “What are the risks facing the electric vehicle industry?” an AI might list high battery costs, charging infrastructure, competition, regulatory changes – covering a broad range that you might not have thought of all at once.

  • Emotionless Analysis: AI doesn’t have feelings. It won’t fall in love with a stock or panic-sell. It just analyzes what’s asked. This can help counter your own biases. If you’re extremely bullish on something, asking the AI for risks might surface facts you overlooked. If you’re too pessimistic, it might highlight strengths you ignored.

  • 24/7 Availability: Markets never sleep somewhere in the world. While you shouldn’t be up at 3 AM stressing, it’s nice that your AI assistant is awake whenever curiosity strikes or news breaks.

🔻 Cons (The Caveats and Dangers):

  • Not Always Accurate: AI can make mistakes or even fabricate answers. This is known as AI “hallucination” – it might claim a fact that sounds legit but is completely false. For instance, people have caught ChatGPT giving wrong financial figures or citing news that didn’t happen. You must double-check critical info. If an AI tells you “XYZ had a revenue of $10 billion last quarter”, verify that from the actual report or a reliable source.

  • Outdated or Limited Data: Depending on the tool, it might not know the latest information. If you’re using a version that’s not connected to the internet, it may not be aware of anything after its last training cut-off. Even those with internet might not access real-time stock prices or certain databases unless specified. Always ensure you’re getting current data for time-sensitive questions (e.g., “What’s the stock price now?” is better answered by your brokerage app or Google Finance).

  • No Secret Sauce: Remember, if an AI model is public, it’s drawing from publicly available information. It’s not getting tomorrow’s news today. Any decent insights it gives (like “Company A looks financially stronger than Company B”) are likely things the market already knows to some degree. You won’t gain a magical edge just by using AI. In fact, if you rely on it blindly, you could become overconfident.

  • Can Encourage Laziness: Investing well requires understanding what you’re doing. If you just take AI output at face value without learning why, you could be in trouble if conditions change. For example, if an AI said “Stock XYZ is a good value” and you don’t know that’s because of, say, low P/E and strong growth – you won’t know how to react if XYZ drops or if some risk emerges. Use AI to assist your judgment, not replace it.

  • Overfitting and Garbage In, Garbage Out: AI can sometimes be too good at finding patterns – even patterns that aren’t real (that’s called overfitting, finding noise and thinking it’s a signal). Also, if your prompt or data is biased or incomplete (garbage in), the answer will be garbage out. Ask a vague question, get a vague (or misleading) answer. As one person noted, how you prompt mattersreddit.com. For instance, bad prompt: “Tell me what stocks to buy.” (The AI might spout some generic list or even apologize it can’t give financial advice.) Better prompt: “I’m interested in technology stocks with strong revenue growth and reasonable prices. Can you suggest a few examples to research further (without giving financial advice) and explain why they might fit?” This will yield a more useful, nuanced response. Always refine your questions for better output.

  • False Sense of Security: It’s easy to be lulled by the confident, oh-so-authoritative tone of AI answers. Don’t forget that it sounds sure even when it’s wrong. As one Reddit user quipped, “AI would rather lie confidently than admit it doesn’t know”reddit.com. So keep that skeptical investor hat on!

Tips to Use AI Investing Tools Wisely

  1. Double-Check Important Info: If the AI tells you a specific stat or “fact” that is crucial to your decision – verify it from an original source. Trust, but verify! For example, confirm numbers from earnings reports, or cross-check news from a reliable news site.

  2. Use Multiple Sources: Don’t just ask one AI in isolation. Try different tools (ChatGPT and Perplexity, for instance) or old-fashioned Google searches. If their answers align, you can be more confident. If they differ, dig deeper to find the truth.

  3. Stay Updated on AI Improvements: The AI tech is evolving rapidly. New models (like OpenAI’s latest or Google’s upgrades) are getting better at accuracy and allowing plugins (e.g., direct connection to stock databases). Keep an eye out for updates – the tool that was “meh” six months ago might be superb now. As one person noted, “It’s the dumbest it’ll ever be today” – meaning it’s only going to get smarter.

  4. Maintain Human Judgment: Use AI as a guide, not a guru. It’s there to assist you in making an informed decision, not to make the decision for you. You are still responsible for your investments. If an AI suggests something that doesn’t sit right with you, trust your gut and common sense. For instance, if it says “This stock is a great buy because it went up 300% last year” – you should think, “Hmm, just because it went up a lot doesn’t automatically make it a great buy now.”

  5. Protect Your Data: If you’re using third-party AI tools, be mindful of what personal info or portfolio details you share. Stick to public data or non-sensitive info when possible, unless you’re sure the service is secure. (For example, don’t paste your account number or entire net worth details into a random AI app.)

  6. Start Small & Practice: If you’re new to using AI, test it out on low-stakes questions or hypothetical scenarios. See where it works well and where it falters. Over time, you’ll get a sense of when to trust it and when to be cautious.

  7. Enjoy the Learning: One underrated benefit – using AI can make researching stocks more fun and interactive. Instead of slogging through dense reports in silence, you have a buddy to chat with about them. This can keep you more engaged and maybe even looking forward to doing research!

The Future: Will AI Replace Investors?

With how fast AI is advancing, you might wonder, “Are we heading to a world where algorithms just trade against algorithms, and humans are out of the loop?” In some ways, that’s already partly true – high-frequency trading algorithms have been around for years, and quant hedge funds use AI models to make trades. Notably, the famous firm Renaissance Technologies uses tons of math and machine learning to drive its Medallion Fund (with eye-popping returns, though it’s very secretive).

However, for everyday investing and long-term investing, AI is more like a co-pilot than a self-driving car. Big institutions might gain small edges using AI, but for retail investors, the value is in augmentation, not automation. Even that AI-run ETF (AIEQ) we mentioned earlier hasn’t cracked the code – it actually underperformed the S&P 500 every single year since it launched in 2017dividend.com. Think about that: a basket of the 500 biggest U.S. companies did better consistently than an AI that tried to pick stocks with IBM Watson’s smarts. That’s humbling for AI hype.

Going forward, expect AI to be embedded in almost every financial service – from robo-advisors that rebalance your portfolio, to smarter chatbots at your bank, to tools that can simulate financial plans in seconds. You might even have AI financial coaches that know your goals and nudge you to stay on track (imagine a Siri or Alexa for personal finance). But will AI replace human investors entirely? Unlikely, at least not in the foreseeable future. There’s a human element to market sentiment, unexpected world events, and personal goals that pure algorithms can struggle with. Plus, as long as humans are involved in markets, understanding human psychology (fear, greed, trends, innovation) remains key – and humans understand humans pretty well.

Bottom line: You don’t have to fear AI taking over your investing. Instead, you can embrace it as a powerful tool in your toolbox. The best outcomes might come from combining AI’s number-crunching and information-processing superpowers with your human creativity, intuition, and judgment.

Wrapping Up: Embrace the AI Revolution – Wisely

AI can be a game-changer for investors in two ways:

  1. Opportunities to invest in the growth of AI itself (potentially reaping big rewards if you choose well and stay patient through ups and downs).

  2. Tools to improve your investing process, helping you research smarter and maybe avoid mistakes.

We’re in the early innings of this AI era – some compare it to being in “the first or second inning of a very long game.” It’s exciting! By reading this guide, you’re already ahead of many in understanding how to navigate it.

To recap, if you want to invest in AI’s future, spread your bets, think about the infrastructure and broad winners, and don’t fall for every shiny AI startup you hear about. If you want to use AI for your investing homework, leverage its strengths (fast reading, explaining, analyzing) but stay alert to its weaknesses (accuracy issues and lack of true foresight). As one wise Redditor said, “If ChatGPT could tell you what stocks to buy, then everybody would be rich.”reddit.com In other words, there’s no shortcut around doing some thinking yourself.

The good news? You now have a new assistant to help think things through. So go ahead – ask ChatGPT to summarize that intimidating 100-page financial report, or have Bing fetch the latest news on a stock you like, or get Bard to help compare two ETFs you’re considering. Just don’t forget to put on your investor cap and critically evaluate what you get.

Investing has always been about staying informed and making decisions under uncertainty. AI can help with the “informed” part, but you still handle the decision part. Use it, enjoy it, but don’t outsource your brain to it. With that balanced approach, you can ride the AI wave rather than be drowned by it.

Happy investing, and may your portfolio be ever in your favor! 🚀🤖📈

Reply

or to participate.