👨‍🚀 Why you shouldn't let new AI models drive you crazy

In partnership with

Hello AInauts,

Welcome to the latest edition of your favorite newsletter! Today, we're taking another spin on the AI model carousel—and why you shouldn't let it drive you crazy.

We'll also show you a new AI shopping assistant and a prompt that will help you solve even the most complex problems in your life.

Here's what we have in store for you today:

• 😵‍💫 Why you shouldn't let new AI models drive you crazy

• 🛍️ New feature: AI decides what we buy (already)

• 💭 Prompt practice: Solving complex problems

Let's go!

Make Every Platform Work for Your Ads

Marketers waste millions on bad creatives.
You don’t have to.

Neurons AI predicts effectiveness in seconds.
Not days. Not weeks.

Test for recall, attention, impact, and more; before a dollar gets spent.

Brands like Google, Facebook, and Coca-Cola already trust it. Neurons clients saw results like +73% CTR, 2x CVR, and +20% brand awareness.

Get a 14-day FREE trial & start improving your ads, at top speed.

😵‍💫 Why you shouldn't let the constant stream of new AI models drive you crazy

The AI world has always been fast-paced. But we admit, the last two weeks have been particularly intense.

First, the improved versions of OpenAI 5.1 in ChatGPT, the new Grok 4.1, and then Google's absolute (albeit overdue) bombshell last week with Gemini 3 & Co.

And you guessed it: the model carousel is already spinning again, and Anthropic has released the new "best" model: Claude Opus 4.5

Of course, we'll dive right in and give you an overview of what it can do. And it's worth reading on. At the end, we'll discuss why you shouldn't let the constant model carousel drive you crazy—and what you should focus on instead.

Gemini 3 is followed by the next big hit: Claude Opus 4.5

We are still completely blown away by all the new Google updates surrounding the new Gemini 3. (We've written a lot about it – here and here).

Google was able to bask in the glory of being the best model for just under a week, but has just had to relinquish the crown to Anthropic's new model.

At least when it comes to programming. But reasoning and problem-solving are also top-notch.

Opus 4.5 is Anthropic's most powerful model. It is particularly well suited for truly complex challenges.

If you've been reading us for a while, you know that Claude is our daily driver. We use Claude the most by far.

We consider it the best of all options when it comes to professional work where mistakes should be kept to a minimum. We are, of course, extremely pleased with the new Opus 4.5 update.

After a day of testing, the new model has already solved Google Apps Script automations that all previous models failed at!

Speaking of work... With the widespread use of deep research, web search, and AI browsers such as OpenAI Atlas or Perplexity Comet, the issue of security is becoming increasingly important.

So-called prompt injection attacks are becoming increasingly dangerous, especially for browsers in which agents work for us. Here, Claude Opus 4.5 is also by far the most stable and least vulnerable.

Why you shouldn't let the constant stream of new AI models drive you crazy

Now let's move on to the most important topic. We see this in many people in the community.

They constantly jump on the next best model. The next hype. Constantly changing tools, etc.

Sure, we're probably not entirely innocent in this and can't keep our hands off it either.

Therefore, we would like to emphasize once again what we believe to be much more important:

• Create clear processes, assistants, and knowledge documents!

• To truly master a tool, to truly work with it every day.

• Use all features and get to know the individual model details.

• ... instead of constantly jumping back and forth.

All major providers and their models are now so powerful that they are very easy to work with.

It doesn't matter whether you use Google Gemini, OpenAI ChatGPT, Anthropic Claude, or xAI Grok.

If you're still undecided, here's our current opinion:

• ChatGPT: Best all-rounder

• Gemini: Perfect for people who work with Google Suite/Tools—and for images

• Claude: Our favorite for professional work, content, and code

• Grok: Real-time updates from X and not-so-strict rules

🛍️ New feature: AI decides what we buy (already)

There is probably no better time than this week to write about this topic.

The world is in absolute Black Friday/Cyber Monday fever. This year, the influence that AI has on our purchasing behavior is greater than ever before.

According to figures from Salesforce, almost a quarter of orders worldwide are already influenced by AI.

This naturally has drastic consequences for retail and existing online structures. We are currently conducting an in-depth analysis to determine what this means for e-commerce companies.

Today, we are particularly interested in the customer perspective. ChatGPT has launched a new feature that makes the shopping process even easier for us:

In addition to deep research for deeper knowledge building, it now also introduces shopping research. With this, ChatGPT aims to become even more of a right-hand man for all your purchasing decisions.

You start it simply with a simple request. ChatGPT then guides you through a few questions to better understand your request.

What's pretty cool is that it also shows you a few products to choose from that you can rate.

Once it has enough information and knows what you want, it gets to work: it searches through all the products and prepares a perfect evaluation for you. Pretty awesome.

Buy directly in ChatGPT

Not quite live yet (only with two test companies) is another feature that will soon be available in ChatGPT: direct purchasing. In just a few seconds.

ChatGPT wants to become your go-to for online shopping. And it will probably succeed.

Of course, the effects are difficult to predict, but we firmly believe that people will always do what is most convenient.

And with your own shopping assistant, you can evaluate hundreds of products based on your preferences and then buy them with just two clicks, which is pretty convenient.

Even if that means we then enter our purchasing decisions into the ChatGPT black box...

💭 Prompt practice: Solving complex problems

This issue is quite long again. So, to conclude, here's something to try out.

Now that your shopping problems have been solved thanks to ChatGPT, you can focus on the really important issues and decisions in life.

Why not try the prompt below? It has helped us a few times already.

(Also ideal for testing models for complex problems, such as Opus 4.5 or Gemini 3 Pro).

Prompt: I am stuck with [problem/decision].
Analyze my situation using the following structure:

1. Problem diagnosis (first principles)
Clarify the actual cause vs. surface symptoms.
Identify limitations, hidden assumptions, and risks.

2. Generate 3–4 strategic options
For each option, specify:
Summary of the approach
Advantages and disadvantages
Risks & potential errors
Required resources (time, money, personnel)
Expected short-term impact
Expected long-term impact

3. Decision matrix (rate each option from 1–10)
Use a weighted evaluation model with:
Impact
Required effort
Reversibility (how easy it is to reverse the decision)
Speed of implementation
Strategic fit
Risk level
Then calculate an overall score and rank the options.

4. Risk map (high/medium/low)
Classify each option in a simple risk map:
High risk/high return
Low risk/medium return
High risk/low return
Low risk/high return

5. Final recommendation
Based on the above data, specify:
The best option
Why it is the best
The first 3 steps I should take in the next 48 hours

6. Bonus: If you see a better option that I haven't considered, suggest it.

Keep your tone objective, strategic, and practical.

You made it to the end—thanks for reading! We’ll be back soon with even more updates.

Reto & Fabian from the AInauts

P.S.: Follow us on social media - it motivates us to keep going 😁!
X, LinkedIn, Facebook, Insta, YouTube, TikTok

Your feedback means the world to us. We read every comment and message—just hit reply and tell us what you think!