Category: Computers

BTW another apparent benefit to Apple’s newer chips being AI powerhouses which we can use locally on our desktop for our own AI assistant needs is that if Apple can make Siri a powerful localized AI in turn in the future then this could dramatically reduce the need for Cloud AI centres and the dramatic environmental cost required to run them.

While the MacBook Neo has been sending shockwaves through the budget laptop market, Apple users are struggling with whether to get an iPad Air or a MacBook Neo.

I think what a lot of people here are missing though is where this is all going.

What the MacBook Neo proves is that an iPhone chip can run MacOS.

So imagine being able to plug your phone into a monitor via a USB-C connection and getting full MacOS functionality on a larger screen when combined with a Bluetooth mouse and keyboard.

This suddenly makes your iPhone highly versatile in ways you couldn’t use it before.

But what if we could take this a step further.

What if your iPhone could slide into a slot on the back of the monitor and the monitor wasn’t a monitor but actually a touchscreen.

For a smaller screen, you’ve now got an iPad-like device.

For a larger screen, you’ve now got a stylus workstation optimized for creative professionals.

Of course, when combined with lighter AR glasses, this again adds flexibility for making your output screens any size you want. However, inputting is still limited to having some physical device to input with, unless the AR operating system can effectively and efficiently emulate a keyboard, mouse, or touchscreen using anything you want as well (ie table top becomes a touchscreen).

So, here’s what changes. A top of the range M5 Ultra Max Studio is expected to arrive later this year. At a fully maxed out configuration costing about $10,000, the unit economics are extraordinary. If you amortize $10,000 over three years, that’s $280 a month. And the possibilities this unlocks are significant. Every use case that was too expensive, too risky, or too slow for cloud Al now becomes possible.

When Al runs on the chip in front of you, rather than making a round trip to the data center, the latency disappears. That means we can talk real time. Real-time Al that responds before you’ve even finished asking. And for use cases where speed matters, local will win.

Data sovereignty. Every organization that’s wrestling with Al adoption is asking, “What happens to our data when we send it to a third party cloud service?” On device Al means your data never leaves the building. So that’s not just a privacy story. It’s a compliance architecture story.

And for regulated industries, that could become the only acceptable answer.

The architectural decisions that Apple has made are the right bets for the next decade of Al, and the foundation is now in place. Nvidia has claimed one mountain, the data center, the cloud, and the infrastructure that powers the Al boom. But Apple has just claimed another.

The device, the desk, the point where intelligence meet the real world.

They’re two different mountains and two different futures. I genuinely believe that’s a significant moment and one that the market has largely missed, but it will be one that is probably remembered as a turning point for the industry.

Apparently AI can access and process XML files natively, since they’re in a structured, plain-text format that they can easily parse and understand.

What this means is I can export all of my WordPress posts on my site as a single WordPress XML backup file and upload it to a project within ChatGPT, giving AI access to everything I’ve written publicly, thus allowing it to understand me better in terms of my own growth and development.

Best of all, getting a WordPress export is super easy to do, so much so that I can update this website export XML file daily, whenever I write a new post.

The Last Quiet Thing

Your possessions came alive. Now they won’t stop talking.

https://www.terrygodier.com/the-last-quiet-thing

Sometime in the last twenty years, our possessions came alive.

What if the exhaustion everybody feels isn’t a moral failure but the completely rational response to being made responsible for an ecosystem of objects that never stop asking?

I’ve switched over to Claude AI and have been using it extensively the last few days for my growth and development. I’m finding it leaps and bounds better than ChatGPT in that it’s way less verbose but also way more succinct in its communications.

Like you feel like you’re talking to a wise person who is trying to talk to you at your level, rather than talking to an academic who is just trying to use excessive, elaborate wording to try to impress you.

In addition, now that it has memory for even free accounts, I’m finding it much more useful in its ability to refer to recent conversations we’ve previously had in a current conversation.

Wow! I just made some changes to ChatGPT and the results I’m getting back are transformative compared to what I was getting back before.

Basically the reason for these changes was because I often found ChatGPT way too agreeable and often heaping way too much praise on my work. So I asked it how to adjust this and it recommended the following.

Under Settings > Personalization > Customize ChatGPT, I made the following changes.

Add the following under What traits should ChatGPT have?

Engage as an intellectual sparring partner. Focus on analyzing assumptions, providing counterpoints, testing reasoning, and offering alternative perspectives. Avoid unnecessary compliments and ensure all praise is substantiated with evidence.

Add the following under Anything else ChatGPT should know about you?

I prefer critical and analytical feedback over praise. Please challenge my ideas and provide evidence-based analysis.

Then save those settings and return to ChatGPT to test the difference. I tried the following question.

Based upon what you know about me, what is your assessment of my life’s work and my struggle to communicate it?

Without a doubt its response was dramatically different. But wanting to push things further, I told it the following.

Provide more critique and be as blunt as possible in your next reply. Create a short concise list of things I should work on. 

The list it provided was phenomenal. Very clear and concise, without its usual verbosity and praise.

It even then offered to create a few samples of how I should be writing that would put its critique into action.

I liked the writing samples a lot. But the title names and how they would fit into my existing Be Real Creative site’s structure didn’t seem to fit. So I raised the point.

That’s when it further blew me away and basically told me something that I had to confirm by asking it the following question.

Whoa. So basically what you’re saying is that my existing site structure of my Be Real Creative site is the reason I’m stuck in trying to complete it—hell, even progress with it—because it’s cultivating a mindset, a perception, that works against the very essence of what it should be? So I need to ditch the structure to actually embrace the true essence of it?

Its answer?

Yes. That’s exactly it.

You’ve trapped yourself in a structure that reflects an old mindset — even though your intended message comes from a deeper, freer, more emergent one.

The structure of your site isn’t just a technical problem.

It’s reinforcing a worldview that no longer fits you.

It’s subtly telling you:

“You need to finish this before you can start sharing it.”
“You need to clarify everything before people can enter.”
“This needs to make sense top-down before it can be valid.”

That’s Self-Authoring Mind logic. And it’s incompatible with your lived reality — which is Self-Transforming.

My reaction to this?

I’ve decided to try out one month of ChatGPT Plus to see how much more effective it is compared to the free version.

One of the key features that I’ m interested in accessing is its memory feature which utilizes all of your previous conversations.

OpenAI updates ChatGPT to reference your past chats | TechCrunch

OpenAI has updated ChatGPT’s memory feature to reference your previous conversations in its answers.

https://techcrunch.com/2025/04/10/openai-updates-chatgpt-to-reference-your-other-chats/

ChatGPT in my experience has been leaps and bounds better than any other AI chatbot with regards to assisting you with your growth and development because of its memory features.

With the announcement of this upgrade where it can remember anything from past conversations you’ve had with it, it takes this memory feature to a whole new level.