- The AI Exchange
- Posts
- AI, Strawberries & the Trough of Disillusionment
AI, Strawberries & the Trough of Disillusionment
3 ways we expect this all to unfold
Welcome to another edition of the best damn newsletter in AI.
As a reminder, we love Reader Questions. Our mission is to make using AI accessible to every business, and a huge part of that is making sure you’re not stuck and that we continue answering diverse questions. So, submit your question here. We’ll review every single one and if we believe i’ll be helpful, we’ll cover answers here in the newsletter.
Here's what we're covering today:
The Trough of Disillusionment
Google’s getting into the custom GPT game
& more
... and if someone forwarded this email to you, thank them 😉, and subscribe here!
Let’s get to it! 👇
TODAY’S PERSPECTIVE
AI, Strawberries & the Trough of Disillusionment
If you’re following AI, chances are you’ve heard of the Gartner’s hype cycle curve. Basically the cycle is after a peak of hype and inflated expectations, we land in the Trough of Disillusionment.
Welcome, we’re probably here.
There could be many reasons for this:
Many announced and not-launched features by major AI players like OpenAI
Rushed AI features launched in nearly every SaaS app known to man
The media cycle over-hyping then over-negging anything that happens
AI is hard — ChatGPT lowered the barrier to entry; but it didn’t mean everything will be easy.
On top of this disillusionment, we have looming launches on the horizon — including OpenAI’s GPT5 (with maybe advanced reasoning and math capabilities, nicknamed project Strawberry).
It’s a weird time.
To provide a little bit of perspective, today we’ll share 3 things that our team’s expecting with future AI improvements:
It still won’t be magic (for a while)
It will be awesome to be a consumer with AI
It will be lucrative to help businesses with AI
It still won’t be magic (for a while)
If ‘magic’ is defined as something that just does what you want without you telling, configuring or controlling it — we think we’re still a while aways from magic.
The biggest difference we expect to come is AI that truly follows the instructions you provide it, reasons through data and shows its thinking, and is able to pay attention to more things and make more novel connections.
It will be awesome to be a consumer with AI
While we won’t have literal magic, it might still feel like we do.
A lot of the best imagineers build consumer apps. And when done right, there can be a lot of money in helping consumers do things — like shop for you, book travel for you, curate education, and more.
It will be lucrative to help businesses with AI
AI is just like any other knowledge work skill.
Just like you can be ‘doing marketing’ without getting results; you can also be ‘doing AI’ with nothing of real value to show.
The value will accrue to the ones who are able to apply AI to make a real impact.
What do you think? What questions are top of mind? Hit reply and let us know!
ANNOUNCEMENTS
Run a marketing agency? You’re Invited!
Join us for our next Free Intro to AI Ops for Agencies Webinar, in partnership with DiviUp
The next one is September 4th at 6pm CST for a crash course on AI Ops designed specifically for agencies! You'll learn how to:
Break down tasks for AI
Create AI playbooks
Automate and elevate your operations
Spots are limited for, reserve yours now!
LINKS
For your reading list 📚
AI is transforming the internet and how we interact with it...
AI is transforming the internet with large language models like Microsoft's Copilot, Google's Gemini, and OpenAI's GPT-4o becoming more accessible to the public.
Google is rolling out custom AI chatbots, called "Gems," for its Gemini subscribers, allowing users to personalize their AI companions.
Google reinstates the ability for users to generate AI-created images of people using its Gemini AI chatbot, with new safeguards to prevent inappropriate or inaccurate content.
AI's impact on image manipulation and the potential risks...
AI image editing tools are no different from Photoshop, or are they? This article critiques the argument and highlights the potential risks of AI image manipulation.
Midjourney, an AI image-generating platform, is expanding into hardware, potentially relating to its ongoing development of AI models for video and 3D generation.
AI's struggles with simple tasks and the ongoing advancements...
Why AI models struggle with tasks as simple as counting the letters in a word, and how this issue is embedded in the very structure of how these models process text.
Regulation and transparency in AI development...
OpenAI, Adobe, and Microsoft are supporting a California bill that mandates the use of watermarks to label AI-generated content.
Anthropic, an AI company, has published the system prompts for its generative AI models, promoting transparency and ethical standards in AI development.
Have a burning AI question? Or stuck trying to help your team adopt AI?
We read every single question submitted and use these to inform the topics we go deep on. 👇️ Submit your question below!
That's all!
We'll see you again soon! Thoughts, feedback and questions are much appreciated - respond here or shoot us a note at [email protected].
... and if someone forwarded this email to you, thank them 😉, and subscribe here!
Cheers,
🪄 The AI Exchange Team