When not to use AI...

Triggered by a post by Nate Jones on Substack (might be paywalled).

I am working on a next big update for SlideMagic, where I am trying to push AI-supported slide generation to the next level. Especially when it comes to layouts with data and text in them, AI is not there yet. The more I start experimenting, the more I realize that AI should not be used everywhere: hallucinations, unpredictable (non-repeatable) outcomes, long wait times, LLM model changes that can completely trip up the tool you have written on top of it.

Nate identifies 4 use cases:

1. Traditional data processing (adding up last month’s sales by product)

2. Traditional machine learning (predicting machine failure base on historical data)

3. Generative AI (summarizing stuff using a chatbot)

4. AI agents (planning your holiday by optimizing and matching lots of constraints)

Work in progress.

More Nano Banana

I worked with Google’s Nano Banana a bit more over the past days, and I think I understands what it is doing under the hood.

“Regular” imaging LLMs predict pixels, you give a prompt, the prompt gets translated into a series of tokens, and the model predicts the best matching pixels given the token input. A flat “soup of pixels” is the result. And because of that, it is hard to make small adjustments to an image, editing one particular aspect and leaving everything else as is.

I suspect Nano Banana works with layers. The model tries to understand what aspect of it refers to the bottom of the pile (the background) and what elements go on top. As a result, it is possible to make very precise edits to individual objects in the overall composition of the image.

In order to make a coherent image, the model needs to have a good understanding of the 3D perspective of the background, and all the objects above it. Like the example about the Porsche in a Dutch town in my previous post, the car gets rotated, and pasted back into the background image with the correct vanishing point in mind.

Vanishing point is preserved when making edits to the image

What the model cannot do is change camera position. view the entire image from a completely different angle. Zooming in and zooming out works. An example is the cover image of this post, where I took an image from the band of my son (Project71) and put them on a big stage. I could not get the model to produce a view from the audience given the image it already produced. (Starting from scratch with an explicit prompt for an audience view would have worked of course).

Note the small glitch in the keyboard of the synth

This is a limitation I can work with for the moment though.

PS. I work with Nano Banana via Google AI Studio, not via its own web site

Nano Banana

I just played around with Google’s “Nano Banana” AI image generator, and it is incredibly good and useful for presentation design.

Current AI image generators take a prompt and predict pixels. Ask for a modification, and a whole new bunch of pixels get generated, redoing the entire image. Nano Banana (we need a better/shorter name), seems to work with layers and objects, and keeps things consistent.

Below 2 quick examples:

“White Porsche in Hoogeveen”

“Turn it around”

Some observations:

• Super fast, the first image was an almost instant response

• Hyper realistic image, does not look cartoonish

• Correct text: the name of the cafe, the license plate, the branding of the car

• (That town looks Dutch, but it is not Hoogeveen)

• But most importantly: isolated editing, changing one thing and leave everything else the same

Photoshop, it was nice meeting you…

I will study the API structure of Nano Banana and see whether I can swap out the image generator in SlideMagic.

Impressive! You can try it out in Google AI Studio

SlideAudit - Academic research to improve slide layouts

I am following academic efforts to use LLMs to improve / automate slide design with great interest. Each takes a slightly different approach. SlideAudit was recently published by Zhuohao Jerry Zhang and others.

SlideAudit teach LLMs what good design is by teaching it rules and principles. A lot of effort goes into building a bank of slides, identifying design flaws for training, synthetically introducing slides, letting the model run and evaluate the results.

I think this approach can work well for publications that resemble print: designs with lots of text in smaller fonts, and images / graphics that are placed in some sort of grid. Books, magazines, newspapers, but also web sites.

Presentation slides are trickier. It is harder to describe what makes a slide a good slide. You know when it when you see a good one (or a bad one), but pinpointing and automating the steps to go from bad to good is tricky.

Visual coincidence

Photographer Ray Giubilo snapped this image at the current US Open Tennis tournament. Searching for “Jasmine Paolini” might come up with this result for the next few years….

View this post on Instagram

A post shared by RAY GIUBILO (@raygiubilo)

It no longer seems to "just work"?

During the Windows - Apple battle of the early 2000s, one of Apple’s major arguments always was that things “just worked”. No drivers to install, software was simple and easy to understand. Things did not crash.

Being both a developer and an amateur electronic musician, I am always a bit late to update my computer’s operating system. I recently upgraded to Sequioa and have noticed a constant trickle of small, little glitches. Difficulties switching audio outputs, Facetime phones that keep on ringing after answering a call, weirdness when waking up a laptop with external monitors attached, apps not working. The solution is usually a machine re-boot.

I hope it is my particular machine set up.

Apple Silicon

SlideMagic has added direct support for Apple Silicon, Apple computers with ‘M” processors. This will make the performance of SlideMagic on modern Apple machines much faster. (Previous version used an on-the-spot translation from Intel to ARM, draining performance).

In the download sections of the web site you will now see 3 buttons: Windows (unchanged), Mac (modern ARM machines), and Mac Intel (older Macs that are still on Intel processors).

The automatic update feature of SlideMagic will not switch you over to the new version. You will have to re-install SlideMagic from the web site. From then on, updates will be for Apple Silicon.

AI images fixed

There was a small glitch in the AI image generator in the SlideMagic desktop app. It has been fixed. Log in and out of your account to get everything working again. (AI images are a pro feature)

AI is good at reading data from charts

Need to make over a slide but don’t have access to the data in a graph? AI to the rescue. Upload a screen shot to an LLM and you get back pretty good estimates of the data values in the chart. It might not be scientifically 100% accurate, but good enough to recreate the graph in your own presentation software.

AI to clean up text tables

One of my big slide puzzles is usually a messy table of pros/cons of a product, or a competitive comparison. How to highlight the right dimensions. Make sure that one is not a sub point of another. Make sure that cells have short text in them. Make sure that text is roughly equally long in each cell.

GPT-5 is very good at this. Copy-paste the messy table into the interface, and the output is pretty useful. Something to integrate in SlideMagic at some stage.

No more disguised examples?

Many confidential presentations often use disguised case examples of clients, potential investments, drugs in development. AI has become incredibly good at uncovering even the most vague ones. Before sending your deck, double-check them in an LLM to see what comes up…

The end of the presentation?

Apologies for the click bait…

I have not been writing here for a while now as I am focusing on 9xc/9vc. But recently, as I am pushing further into the world of AI, all my pas experiences seems to be coming together: computer science, company analysis, presentation design, and hardcore biopharma science…So I might occasionally come back here.

The majority of “presentations” are documents that are used make decisions inside companies. They happen to have graphs and other visuals inside them, hence the word “presentation”. Most humans are not very skilled in writing a concise memo to make a point: hence bullet points in large font and visuals to the rescue.

AI could change that: boiling down these big slide decks to a few paragraphs with a decision that needs to be taken, pros/cons of alternatives, and the recommended way forward. No slides needed.

Slides -> Text -> Slides

Most project presentations consist of slides ripped from the project working documents. Pages you used to learn and discover things to analyze and make conclusions. They are not the best visuals to communicate the project outcomes to outsiders.

• Their layout is likely to be messy and detailed, designed to be read at close distance

• The structure of your project document is likely to be methodological, like a workplan that makes sure you cover all the bits of work required.

Here is an alternative approach to make your final presentation:

1. Leave your project working papers for what they are

2. Write a long-hand text (1 page) that explains what the conclusions and next steps are, supported only by facts that are critical in supporting the recommendations. (Not a “this is what we did” paper)

3. Now, make a completely new slide deck based on just that text document

The crop of Trump's post attack picture

This image will go in the history books anytime the election of 2024 will be discussed. Image credit: Evan Vucci , image analysis: David Altizer

The trick here is cropping. A good photographer will have a good first start when snapping the image, but adjusting things slightly afterwards can add a lot.

Stock image 2.0

Stock images can be cheesy and cliche, and the current AI image generators are trained on tons of stock images. So if you ask for an image using a stock-image-like prompt, you will get a stock image…

Stock images can happen to the best of us, above, SlideMagic had a go at “a happy group of young adults taking selfies”.

Quick makeover: our high-tech industry matters

I saw the following slide coming by in my Twitter feed (original post):

Here are some things I fixed:

• Message in the title

• 19.7% -> 20%

• Simplification of the labels

(For those of you interested in the political context: Israel is fragmented in many population groups, the Israeli high-tech sector which is mostly secular, pro-democracy, situated around Tel Aviv, and not really represented in the current government is making the case that it generates a big chunk of the Israeli economy and is bank rolling many other sectors.)

More server updates.... Might need to log out / log in

I am making updates to the server plumbing of SlideMagic, to ensure uninterrupted service, please log out and back in to the app. Apologies for the inconvenience, but this keeps everything safe and secure.

Version 3.1.8.

I deployed a new version of the SlideMagic app today with security updates. The server was patched too. Please let me know if you experience any issues.

It's just a draft

When quickly putting a draft presentation together, it is tempting to not spend any effort at all on design and layout. “We can always fix that later”.

I would argue the opposite. Make the draft look as good as the final product will be. It sets the entire mood for the project. Looking at messy / ugly charts is not a big motivation to do great work. Messy / ugly charts encourage people to add bullet point and too much text, because you can always remove it later.

The good news is that a simple chart with simple content does not take a lot of effort to design properly. Fix proportions, alignment and colors and everything looks great in a few clicks.

(Pro-tip: use SlideMagic for your draft documents)

Data chart labels: inside or outside?

This post on the F1 Instagram account has an interesting solution to the labels of a data chart:

• When the column is tall enough, the label is put inside with a contrasting white font color

• If not, the label is put on top, if possible with a font color that matches the column

The advantage is that this enables you to make the column graph taller as a whole (no need to budget extra space for the labels), and it is easer to put a background image as most labels (in tall columns) have enough contrast against them.

The vertical orientation of the labels though…