Newsletter Subject

This Is The Newest, Biggest Thing That Will Make Your Life Easier

From

bakadesuyo.com

Email Address

eric@bakadesuyo.com

Sent On

Mon, Feb 19, 2024 11:02 AM

Email Preheader Text

Barking Up The Wrong Tree February 19th, 2024 ------------------------------------------------------

Barking Up The Wrong Tree February 19th, 2024 --------------------------------------------------------------- Before we commence with the festivities, I wanted to thank everyone for helping my new book become a bestseller! To check it out, click [here](. --------------------------------------------------------------- This Is The Newest, Biggest Thing That Will Make Your Life Easier ([Click here]( to read on the blog) Up until very recently, most of our experiences with AI boiled down to shouting "ALEXA, STOP!" AI means “artificial intelligence”? More like “Awkwardly Inept”, if you ask me. But things have changed... Enter ChatGPT, strutting onto the scene like a rock star. Suddenly, AI wasn't just a voice in a box that occasionally understood your mumblings; it was like having a chat with someone who'd actually read a book. Or, you know, all of them. How well could a cold, heartless machine understand the profound beauty of the written word, the subtle nuances of human emotion, the delicate dance of grammar and punctuation? Uh… turns out, pretty well. GPT-4 [routinely beat ~80% of humans]( on most standardized tests, including the SAT, the GRE, the LSAT and the bar exam. Another [study]( had people write business documents. Leveraging ChatGPT made them 40% faster and the results were 18% better. Better yet, those who had ChatGPT assistance said they were happier after. It made boring work easier. Artificial Intelligence is now legit. You have a very smart intern to do your bidding 24-7. And it never gets bored. Now a lot of people say, “Oh, I tried ChatGPT and it wasn’t that good.” This response is understandable. Getting started with ChatGPT is like being given a magic wand without the spellbook. It seems exciting and promising, but mostly you just find yourself waving it around, hoping for the best. So it’s time for a basic breakdown of how ChatGPT works and how it can work for you. This won’t be exhaustive. It can’t be – the technology is too new. This may all be outdated very soon but the goal is to give you a primer. (And, no, we’re not going to wade into the Skynet / AI-will-kill-us-all debate. You can get plenty of that elsewhere.) This is a big, ever-growing topic so we’re going to focus on [ChatGPT]( which is currently the best AI of the bunch. The free version is powered by GPT-3.5. But GPT-4 is much better. You can either pay for a subscription or access GPT-4 free by using Microsoft Copilot [here](. I want to acknowledge the fantastic work of Wharton professor Ethan Mollick, who I’ve learned a lot from. His [blog]( is definitely worth checking out. The rest of this is drawn from my own trial and error… and more error, and even more error. Let’s get to it... What Is ChatGPT? GPT stands for “Generative Pretrained Transformer”, which sounds like a part you'd need to fix a broken time machine. It’s the tech under the hood that drives ChatGPT. To criminally oversimplify, ChatGPT is basically a mega-brainy AI that's swallowed all the information on the internet and is now ready to burp at your command. It all started with a 2017 academic paper, "[Attention is All You Need]( That title might sound like a self-help book for narcissists but the big idea it had was that by focusing on (or paying "attention" to) what's important in a sentence, ChatGPT can predict what should come next with eerie accuracy. This system of “next word prediction” sounds rather simple but turns out to be insanely powerful. So the developers fed it a gargantuan amount of information and the results were staggering. Imagine someone took every book from Amazon, threw them into a blender, and said, "Make me a smoothie that can talk." A frequent criticism of the technology is that it occasionally “hallucinates” – makes up inaccurate stuff. And that definitely happens. But instead of focusing on the fact that it sometimes gets things wrong, ask what your point of comparison is. People and the news get a lot of stuff wrong too. So if you're looking for heart surgery advice, consult a professional. The good news is ChatGPT has access to the internet, so you can have it verify its own responses. Now people who don’t know much about the technology might be wondering, “If this thing isn’t reliable, why use it to search for answers?” And my response would be: if you’re only using it to search for answers, you’re using it wrong... What ChatGPT Is Really Good For A lot of people use it as a search engine and that’s fine. But thinking it’s primary use is as a Google replacement is a fundamental misunderstanding of the technology. ChatGPT isn’t for searching – it’s for doing. Ask yourself: “If I had a smart assistant working remotely, what might I ask them to do?” Now that your mind has gone blank, let’s drill down on some specific use cases: 1) Writing Putting together a document for work? Getting started can be hard. Ever found yourself staring at your computer screen, your brain as blank as the document in front of you? Your mind is like, "Words? What are those?" ChatGPT can churn out rough drafts for work faster than you can say, "Please, no more team-building exercises." Reports, emails, presentations, resumes, cover letters, etc. And you can give it pretty subtle instructions to customize things: "Make me sound smart but not pretentious, professional but not robotic, casual but not like I'm texting my high school BFF." Need ideas for a project? It's like having a brainstorming buddy who's had six espressos and read the entire internet. Trying to make a decision? Ask it: "I am trying to decide if I should [X]. Give me a list of pros and cons.” Need research assistance? It's like having a tiny, know-it-all librarian in your pocket, minus the shushing. You can give it your own writing and have it make improvements. It can check grammar, make your writing more professional or you can even ask it to rework something in the style of a particular author. (As a writer, doing a blog post about GPT can feel like crafting your own obituary.) 2) Summarizing Give ChatGPT a URL or upload a PDF and you can ask for an overview at any designated length. Is it a complex topic that makes your head hurt? “Explain this to me like I'm 11 years old.” And after the summary you can ask questions about the document. “What does it say about X? What is the most important part of this document? What about this is new?” 3) Recommendations It can recommend books, movies and music based on your preferences. “I like X, Y, and Z. Can you suggest similar ones?” And you can be insanely specific: “Give me the best movies between 1955 and 1965 that are not related to war.” 4) A Mentor Do I even need to explain how helpful this can be for students? ChatGPT is like that nerdy friend who actually did the reading for class. But ChatGPT can be a fantastic learning amplifier for anyone. It can interactively teach you a foreign language. Ethan Mollick uses it to help students improve their [negotiation skills](. Or get ChatGPT to help you prepare for a job interview: “Let’s role play a job interview. After I respond to each of your questions, tell me how to improve my answers.” 5) Projects Ask it for a recipe based on the ingredients in your fridge, and voila! You've got a meal that won't trigger your smoke alarm. Upload a photo of your freshly baked sourdough bread and it’ll give suggestions on how to improve it. It's like having Gordon Ramsay in your phone, but without the screaming and expletives. ChatGPT is a great helper for learning new hobbies, DIY project assistance, gardening tips, fitness routines, travel planning, and on and on. Mollick has a [post]( on how it can help you develop an idea for a new business. Okay, enough big picture. What are the first steps to using it well? How To Get Started With ChatGPT People usually give up too fast when they first start playing with it. They stare at the blank screen, finally ask it something bland... and get a bland answer. Then they go, “Meh” and quit. First: make sure you’re using GPT-4, whether it’s through a paid subscription or free on Bing. GPT-4 is just plain smarter. Second: go in with real intent. Look at the above list of things we saw it can do and pick something you actually want help with. A real problem you want to solve. And the third big issue is “prompting.” Basically, how you talk to it. You can type conversationally, but there are tricks to get better results. Remember, this software knows a LOT of things and can’t read your mind. Here are some tips for writing better prompts. These are by no means essential, but they’ll usually get you better results: Tell It Who It Is Giving it a relevant persona can help you get the results you want: “You are a marketing expert.” Tell It Who You Are Asking it to help you design an exercise program is great but if you don’t tell it whether you’re young or old, trying to lose weight or gain muscle, well, you’re going to get something very bland and cookie cutter. Tell it who you are and your related goals to get a more focused, relevant response: “Write this PowerPoint to be understood by someone who is college educated but has no formal business training.” Be Specific About What You Want Be too specific when doing a Google search and you get few or no results. ChatGPT is the opposite. It knows so much that if you ask generic questions, you get generic answers. More constraints give better results. Don’t say, "Tell me about dogs." Say, "Give me 10 progressive, chronological training tips for a Labrador puppy. I have never trained a dog before.” And think about what kind of response you want. A rant, an essay, an explanation, and a report are very different. Want a bullet point list? Maybe a step-by-step guide? Should the answer be concise or exhaustive? Do you want it in the style of Wikipedia -- or Chris Rock? And tell it what you don’t want: “I’m already familiar with X so you don’t need to cover that. Focus on Y.” Long Prompts Are Good. Give It Context And Vibe. Longer prompts are better. Using more details and points of reference is powerful. It’s the difference between dropping an old-fashioned bomb versus a GPS-guided drone strike. Showing your expertise on a topic will help it understand the level of response it should give. A friend of mine who’s a nurse gave it the prompt: “Please provide specialized dressing choices for 70-year-old with skin ulceration on tailbone due to mechanical injury. Wound bed is moist with minimal exudate.” Using that technical terminology “told” ChatGPT she’s a pro and it should answer like a pro. What if you’re not an expert? Providing examples of the kind of response you want can really help. Give it an idea of what the bullseye looks like so it can better hit it. And if you’re unsure of what you’re looking for you can even just give it vibes: “The essay you generate should have the calming feeling of watching a beautiful sunset.” Iterate Another reason people give up too quickly with ChatGPT is they think they’re going to put in a brief prompt and immediately get a perfect result. Again, it can’t read your mind and you likely need to give it more context. Don’t see it as a one-shot question and answer system. Treat it like a conversation. Go back and forth, guiding it toward a result that’s closer and closer to what you want. Seeing this as a collaborative process is powerful. Remember: it doesn’t get bored. Ask it for ten iterations. Then ten more. And give it feedback each time: “Again but more creative and vivid.” Okay, time to round it all up and give you the secret to getting better with ChatGPT... Sum Up Here’s how to improve your life with ChatGPT: - What Is ChatGPT?: You now have a remote, super-smart staff to help you with almost anything. ChatGPT is like an enormous pile of books that's good at pretending to be human. Which, honestly, sounds a lot like me in high school. - What Is It Good For?: It’s not just a new search engine. It can write, summarize, teach, make recommendations, or assist with almost any project… and that’s just for starters. Be creative. (When it’s done you can even ask it to write a limerick to help you remember important takeaways.) - How Do You Get Started?: Pick a real problem you want help with. Tell it who it is, tell it who you are, be specific, use detailed prompts, and iterate. Using short, vague prompts is like telling a chef “make food” and being shocked when you get a dish that isn’t what you wanted. Your Jedi training is not complete. (And, no, I am not available for tech support.) In some ways, ChatGPT is less like a computer and more like casting magic spells. Nobody fully understands how the technology works, but as you practice you get better at whispering the right incantations to get your desired result. So put in some time. Practice. Experiment. I swear, you’re not going to break it. Some will reply, “Ugh, that sounds like a lot of work.” It’s not. In fact, it’s play. The secret to getting amazing things out of ChatGPT is curiosity. Prod, poke, and question like a child who's just discovered the word "why." You can now make amazing things you never had the power to create before. Missing out on this would be like refusing to use the internet back in the '90s because you thought it was just a fad. Imagine future generations, looking back at this era, shaking their heads in disbelief that some chose to ignore the AI revolution because they were too busy scrolling through pictures of other people’s brunches. So, revel in this new, AI-infused reality. Dive in. Because in a world where a computer program can write poetry, create images, and maybe -- just maybe -- understand the human heart, the only thing more insane than the questions we ask would be not asking any at all. ***If you are one of those lovely people who bought "Plays Well With Others" please leave a review on Amazon [here](. Thanks!*** Email Extras Findings from around the internet... + Want to know how to stop assuming the worst will happen? Click [here](. + Want to know how to improve your friendships? Click [here](. + Want to know how your voice can increase your social status? Click [here](. + Miss my prior post? Here you go: [4 Secrets Of High-Performing Teams](. + Want to know what predicts relationship satisfaction? Click [here](. + You read to the end of the email. Whew. That means ChatGPT hasn't replaced me yet. I'll sleep better tonight thanks to you. Oh yeah -- it's Crackerjack time: Data and technology are changing the game -- many games, actually. And the very smart Seth Stephens-Davidowitz has used AI to break down some of the most interesting insights about basketball. If you're curious about what AI can tell us about sports, check out his new book [here](. Thanks for reading! Eric PS: If a friend forwarded this to you, you can sign up to get the weekly email yourself [here](. This email was sent to {EMAIL} [why did I get this?](    [unsubscribe from this list](    [update subscription preferences]( Bakadesuyo · 8033 Sunset Boulevard, #1073 · Los Angeles, CA 90046 · USA

EDM Keywords (293)

young yet wrong writing writer write would worst world work word wondering without wikipedia whispering whether well waving watching wanted want wade voila voice vibes verify using use url upload unsure unsubscribe understood understand turns trying trigger tricks trial toward topic tips time thought thinking think things thing thanks texting ten tell technology tech talk system swear swallowed summary subscription style step starters started staring stare spellbook specific soon someone solve smoothie sign shushing shocked sent see secret searching search screaming say saw sat round review revel results result rest responses response respond report replaced reliable related reference recently ready reading read rant quickly questions put pros promising project professional pro primer pretending prepare preferences predict practice powerpoint powerful powered power post points point play pictures photo phone people pdf part overview outdated opposite one new never need narcissists mumblings much mostly mollick moist missing miss mine mind might mentor meal maybe may makes make lsat lot looking list limerick like life librarian level legit learned knows know kind internet instead insane ingredients information increase improve important ignore idea humans human hood helping helpful help heads happier great gre grammar gpt got good going goal giving given give get generate front friend fridge free focusing focus fix fine find festivities feedback fast fact explanation explain expertise experiences exhaustive even essay error end email elsewhere dropping drill drawn done dog document dish discovered disbelief difference develop details design decide debate currently curious creative create cover context concise computer complete comparison commence command closer click class churn chose child check chatgpt chat changing burp bunch brunches break brain box books book blog blender blank bland bestseller best basketball basically available assist asking ask around anyone answers answer amazon almost ai actually acknowledge access 90s 1965 1955

Marketing emails from bakadesuyo.com

View More
Sent On

13/05/2024

Sent On

29/04/2024

Sent On

15/04/2024

Sent On

01/04/2024

Sent On

18/03/2024

Sent On

04/03/2024

Email Content Statistics

Subscribe Now

Subject Line Length

Data shows that subject lines with 6 to 10 words generated 21 percent higher open rate.

Subscribe Now

Average in this category

Subscribe Now

Number of Words

The more words in the content, the more time the user will need to spend reading. Get straight to the point with catchy short phrases and interesting photos and graphics.

Subscribe Now

Average in this category

Subscribe Now

Number of Images

More images or large images might cause the email to load slower. Aim for a balance of words and images.

Subscribe Now

Average in this category

Subscribe Now

Time to Read

Longer reading time requires more attention and patience from users. Aim for short phrases and catchy keywords.

Subscribe Now

Average in this category

Subscribe Now

Predicted open rate

Subscribe Now

Spam Score

Spam score is determined by a large number of checks performed on the content of the email. For the best delivery results, it is advised to lower your spam score as much as possible.

Subscribe Now

Flesch reading score

Flesch reading score measures how complex a text is. The lower the score, the more difficult the text is to read. The Flesch readability score uses the average length of your sentences (measured by the number of words) and the average number of syllables per word in an equation to calculate the reading ease. Text with a very high Flesch reading ease score (about 100) is straightforward and easy to read, with short sentences and no words of more than two syllables. Usually, a reading ease score of 60-70 is considered acceptable/normal for web copy.

Subscribe Now

Technologies

What powers this email? Every email we receive is parsed to determine the sending ESP and any additional email technologies used.

Subscribe Now

Email Size (not include images)

Font Used

No. Font Name
Subscribe Now

Copyright © 2019–2024 SimilarMail.