Vertiport GPT Assistant

OpenAI and ChatGPT is dominating the LLM space for more than a year now, at least in media and public attention.

AI generated eVTOL design (Midjourney)

While ChatGPT 3.5 was already amazing, version 4.0 is a major step forward. GPT-4 comes with several notable features and capabilities that represent significant advancements over its predecessor. With supposedly 100 trillion parameters, it would surpass 3.5 500 times.

Released in March 2023, GPT-4 shows a better grasp of context, allowing for more relevant and coherent responses. It maintains coherence over longer passages, making it suitable for generating lengthy reports, stories, or detailed explanations. Trained on a more recent dataset, GPT-4 has access to newer information, making its responses more current. It improved in handling complex reasoning tasks, problem-solving, and analytical thinking and exhibits a lower propensity for generating incorrect or nonsensical responses, particularly in nuanced or complex subject areas.

Version 4 is solely available through subscription of currently at a cost of $20 per month.

Now we have the possibility to add knowledge to the model and make it accessible for private or public usage.

Let’s try a hands-on example, building a Vertiport Assistant to answer around eVTOLS and vertiports. It has already common knowledge about the topic.

Now let’s ask something that OpenAI has not scraped, like the FAA EB 105, the Engineering Brief that describes the design of vertiports. The document was released in March 2023 (GPT 4 covers information until April 2023) and is publically accessible at this link.

ChatGPT does not know about it, only provides some indication recognizing the meaning of FAA and EB.

Lets create a GPT and add the PDF file to it.

Note, I also added the EASA equivalent document, the PTS for Vertiports (March 2022) as well disabled web browsing to make sure we rely on uploaded knowledge only.

Now we receive an answer which is more or less a rephrased version of the document.

Asking for more specific informations results in deviations.

We are off by 5mm ?!

Conclusion: With the current version 4 we can integrate ChatGPT easily into other solutions, using its multimodal features and inject our own knowledge. Still, we have to be careful to rely on specific information and values.

Debunking Financial Scam Ads

Update Jan 19th 2024: Same content under different web address. See below.

Quite likely the internet as we know it today would not be able to exist without ads. Advertising is as old as media content is published and financed. We grew up in a world (years back) where the brands and companies behind the ads in newspaper and TV were more or less known to us. Let’s face it, today’s online ads are throwing a lot of rubbish products and scams at us.

YouTube is only possible becaue of ads. Do you notice the increasing number of ads and length shown in almost any YouTube video recently? Less and less you can skip and the 10sec ads increased to 20sec and more without a chance to run away from it. But, to be fair, it is a free service, so bear with the ads (to some extend).

^{AI generated image}

BUT the amount if scam is unbelievable and Google seems to do little to cut down or stop it.

As a little academic exercise, let’s debunk an ad.

In this video ad, interrupting another video I was watching, Elon Musk is talking in German language (voice over, quite likely by an audio AI service, the voice and tone is very similar to his real voice) about financial investment in Tesla and making a thousand euro every day without doing anything.

Whats wrong ?

• The original video is recorded at an AI conference 3 years back. Elon is not talking about investment at all during this presentation, but solely about autonomous systems. The subtitles are cut off in the ad version above.
  Reference: https://www.youtube.com/watch?v=SHOLLbFZXeE

• The ad is pointing to a website ending with the cfd toplevel domain, which is meant for companies in the clothes, fashion and design industry. CFD can be mistaken in this context for the financial investment term “contract for differences”.
• The website is showing the Tesla logo, has zero indication about the company behind it, no “about us”, terms of use, contact, etc. but asking for your contact and phone number.

• Plenty of successful customer references are displayed.

• Several names with the same profile image. The same images are used on other similar websites.
• There is no traces of the company name (used in the domain name) in the web.

Conclusion: The multi-billion dollar organization Google can spot copyright and IP infringment in minutes after uploading regular content but cannot take down obvious scams ?
You can report these ads, but new ones will re-appear every minute. Ignore them and press skip as soon as possible.

Update Jan 16th 2024

The exact same video on Youtube.

A new website with similar content, this time also using the German government.

This time starting to use AI generated faces for their reference customers.

Update Jan 19th 2024

Video footgae taken from the interview with Elon Musk in 2022. Putting different text into the audio.

Reference: https://www.youtube.com/watch?v=2WX_mgnAFA0

Another website harvesting for your contact information.

Again, success stories with fake people and pictures, borrowed fro somewhere else.

Rainer Bade: The pictures shows investor Matthew DeBord

Disclaimer: This research is for educational and academic purpose, no copyright infringement intended.

Inter-AI Conversation

Again, there are so many new AI based services starting up every day and we still have to find the ones useful and really adding value, like an assisting AI service. Though there are also obscure and funny applications.

AI Generated: Robot texting another robot

This is a very brief example of an absurd use case from the world of text creation. We want to create a summary or overview text about a certain topic, in order to equip an article or paper with content.

Lets try the following in the context of aviation and sustainability:

Create a half page summary about the following bullet points:
- The future of aviation and sustainability 
- The use of SAF in aviation 
- Use of hydrogen in aviation 
- Use of battery technology and electric propulsion in aviation

ChatGPT’s text proposal:

The reader of our document does not want to spend the time reading this and asks ChatGPT to summarize the text that was created by ChatGPT.

Conclusion: no words.

Even more Airport Self-Service-Kiosk Designs in 2024

Another 6 months later (December 2023) and Midjourney released 6.0 as alpha for image creation, worth another visit to same and new design prompts. We are going more and more photorealistic without any additional prompt attributes. Some of the design are little practical I guess but certainly create a decorative element in the terminal.

Previous posts: How to design an Airport Self-Service-Kiosk (Jan 2023) and More Airport Self-Service-Kiosk Designs (Aug 2023)

Disclaimer: This is solely for educational purpose and does not reflect any actual products or design I am involved with (until today). No copyright infringement is intended.

Style of 1900s

Style of 1910s

Style of 1920s

Style of 1930s

Style of 1940s

Style of 1950s

Style of 1970s

Style of 1980s

Fantasy Dreamland Style

Mad Max Movie Style

Art Deco Style

Rococo Style

Gothic Style

Lord of the Rings Style

Kids Toy Style

Apple Design

Andy Warhol

Google Design

Neon Design

Joan Miro Design

Design by Pablo Picasso

Hieronymus Bosch Design

Salvador Dalí Design

Colani Design

Zaha Hadid Design

Conclusion: Runing Midjourney V6.0 creates very photorealistic images, that could easily been taken for a real photograph. It is mindboggling how much it advanced in one year. The challenge, create new unique designs that are not just copying the stykle of someone else. Though creativity contains inspiration and copyying.

Debunking AI generated Content and Fake Accounts

Update 2024-01-19: Update section at the end with more debunked content.

Not one day without yet another announcement of some new AI tool or platform that creates content, like text, images, audio, video, sourcecide, etc. You see new tools coming and going. Some are here to stay. Some are useful, some are questionable.
More annoying is the increased volume of content that is polluting social media or the professional platform LinkedIn.

AI Generated image showing a printer throwing out posts.

I came across this posting, one random smaple of many, talking about the most lucrative careers in the world. Let’s dissect this piece of random click-bait AI created postings.

LinkedIn Posting 1

• Off topic: I posted it in Angular group. Reason to leave these groups if not moderated.
• The author is working for a marketing company, no surprise. The only activities are postings that lead to similar articles.
• The post is hosted on blogger.com, the free blogging resource by Google/Alphabet. Financed by Google AdSense advertising.
• The same LinkedIn post is published by various other authors all working for similar companies. Most likely these persons are either non-existing and just fake accounts, or just low paid mechanical turks (Amazon crowdsourcing platform).
  You cant mute these accounts fast enough because new ones getting created and they start posting in the same groups. Only choice, leave the group.
• The image is an unrelated random lead picture, usually taken from free image repositories, stock images or more frequently now, AI generated. In this case we see the Rose Bowl stadium during the 1994 World Finals, the picture showed up on reddit a few days back and cannot be found in tineye.

..continued below.

LinkedIn Posting 2

LinkedIn Posting 3

The above links leads to the blogspot post below.

Screenshot of the blogger hosted blog/post.

The whole blog contains nothing but these kinds of posts with a lot of SEO targeted labels to attract more clicks.

Tool detecting most text as being AI created.

(Reference: https://gptzero.me/)

Conclusion: It becomes harder to navigate the content landscape and find/confirm human created articles and images. Maybe social media sites and professional sites should have a tick logo for real content, I expect this feature soon. Or, in reverse, some platforms will ban AI content all together.

What is going to happen in future if less and less content is human created ? AI is not genuinly creative and we will see similar content again and again.

Advertising Spam on LinkedIn

The second category of content that polluts profesional groups and makes it hard to find relevant content, are marketing companies advertising agressively for their expensive market reports.

A good sample is this one, this account is bombarding various aviation related groups with advertising.

Located in the U.S., working for a company in Germany that does not exist.

The profile picture is all over the internet for fashion related websites.

The account activities are nothing but postings for reports about various fields in aviation and military. They all link to expensive reports or research reports.

Some of the research papers covers:
𝐀𝐯𝐢𝐚𝐭𝐢𝐨𝐧 𝐂𝐫𝐞𝐰 𝐌𝐚𝐧𝐚𝐠𝐞𝐦𝐞𝐧𝐭 𝐒𝐲𝐬𝐭𝐞𝐦 𝐈𝐧𝐝𝐮𝐬𝐭𝐫𝐲, 𝐆𝐥𝐨𝐛𝐚𝐥 𝐍𝐚𝐯𝐢𝐠𝐚𝐭𝐢𝐨𝐧 𝐒𝐚𝐭𝐞𝐥𝐥𝐢𝐭𝐞 𝐒𝐲𝐬𝐭𝐞𝐦 (𝐆𝐍𝐒𝐒) 𝐈𝐧𝐝𝐮𝐬𝐭𝐫𝐲, 𝐓𝐡𝐞 𝐅𝐮𝐭𝐮𝐫𝐞 𝐨𝐟 𝐂𝐚𝐫𝐠𝐨 𝐒𝐡𝐢𝐩𝐩𝐢𝐧𝐠 𝐈𝐧𝐝𝐮𝐬𝐭𝐫𝐲, 𝐀𝐢𝐫𝐜𝐫𝐚𝐟𝐭 𝐋𝐞𝐚𝐬𝐢𝐧𝐠 𝐈𝐧𝐝𝐮𝐬𝐭𝐫𝐲 and many more.

More Fake Content and Scam

Update 2024-01-19

LinkedIn post about guide to turbofan engines. The text is 100% AI generated and is a click-bait to a page selling reports at ridicolous prices.

GenAi Imagery Increasing

No doubts, since 2022 the generative AI space was exploding. It left the research stage and became mainstream in 2023. It is getting harder and harder – at least for the human brain – to spot genuinely created content (images, videos, text, audio) nowadays. You won’t even know if this blog post was created by me reflecting about the topic and actually writing the text or just me throwing a text prompt at some text generation service and copying it here.

It started with first attempts to create images with Generative Adversarial Networks (GAN) in 2014 (Ian Goodfellow, University of Montreal). All made possible with the advent of deep learning and (convolutional) neural networks utilizing massive CPU (GPU) processing power and a vast amount of data available today. Most people do not understand what is going on under the hood of all this or grasp the concept, except you are really into computer science and ai. The results are not transparent at all, eventually it will improve due to regulation and legislation (to be proven!).

I enjoy using the tools and a couple of my blog posts talk about GenAI or using the tools for content creation, mostly images using tools like DALL-E or Midjourney or local Stable Diffusion.
I believe I (still?) can distinguish and spot generated images, especially the ones that are used unaltered out of the image generator, mostly with simple text prompts. More and more images in my LinkedIn stream (presented to me) are artificially created, no problem with that if it serves a decorative purpose as a design element, the problem starts when the images is supposed to document facts and actual events or people.

A few hints how to spot these images, using an actual, more or less random, website as example. I am not picking on the content or the authors, I just highlight how you can identify generated images. Let’s have a look at this website about travel and mobility technology (TNMT). I came across this article about passenger frustration in my LinkedIn stream, the lead image grabbed my attention (doing a good job because that’s its purpose). See below screenshot of the post.

At first sight it looks real, a passenger in a crowd or queue, a frustrating scene (delay, cancellation,..). The primary subject is very photorealistic, but the background reveals the source not being a real photo shot. (Quiten often it is the background that pinches me).

Usually generated images are not highlighted as such and there is no mention of the photographer or the image source.

Some artefacts in the image that support the suspicion:

The right eye of the person looks unnatural or like someone coming out of a box fight.

Missing teeth or need to visit a dentist.

Unreadable text on the t-shirt.

Out of scale ear of the person next to the subject.

Person with green jacket, arm point forward and either no head or the head facing backwards.

Another important tool, is the image reverse search. Check if the image is used anywhere else. Last time editorial images were dominated by stock images and you get dozens of references, but not for generated images. Why bother to copy/steal a generated image if you can create a new one with 5 mouse-clicks. You can use TinEye or the respective Google service. Above image:

TinEye result

Google image search result

I recommend doing this if you doubt the source of images in the news, especially coming from social media platforms.

Additionally, you can try to use online tools to identify generated images, but be aware, you will get very different responses. Some tests in no particular order:

Hive Moderation

Content at Scale

AI or Not

Let’s dissect a few more images from the same website. It seems they started to use generated images instead of stock images in Nov 2022. Before that time all images were stock images.

• Unreadable text
• Some kind of fantasy engine

• Missing landing gear
• Broken aircraft physics.

• Frankenstein plugs
• Floating and missing engine
• Unreadable text

• Obvious

My conclusion and recommendation: Feel free to use generated images for decorative or editorial purposes. Maybe you should mark them as being generated, at least somewhere in the text. Avoid image rubbish and obvious artefacts.

More Airport Self-Service-Kiosk Designs

6 months later, the second round using the Midjourney 5.x release for image creation, worth another visit to same and new design prompts. Some of the design are little practical I guess but certainly create a decorative element in the terminal.

Disclaimer: This is solely for educational purpose and does not reflect any actual products or design I am involved with (until today). No copyright infringement is intended.

Technical Drawing in the 1970’s

1900’s

1910’s

1920’s

1930’s

1950’s

1970’s

1980’s

Art Deco

Romanesque Era

Renaissance

Caveman Style

Futurama Style

Simpsons Style

Rococo Style

Surrealism

Joan Miro

Pablo Picasso

Andy Warhol

Legoland

Playmobil

Fantasy Dreamland

Neon

Mad Max

Running image generator on your local machine

The hype around ChatGPT was going through all media over the last few months, seeing all kind of statements from “we are all losing our jobs” to “General AI and singularity is near”.

Now the generated images are picking up quickly in the news because of fake image being released showing celebrities and other public persons in situations that are very misleading basically. Going to the extend that Midjourney stopped free trials.

More protective filters getting added quickly to prevent people from creating images with certain context, like stopping you from entering most celebrity and people in politics.

To create images using nothing but a text prompt and without any datascience or AI background, you can go to the various platforms that offer the image creation as a service.

Some of the models are opensource (stable diffusion) some are propietary (DALL-E). It is possible to run the model on a decent equipped local machine. Let’s try that.

lexica.art

beta.dreamstudio.ai

labs.openai.com

www.midjourney.com

Tutorial

How to run this on a local machine without any limitations and at no cost (certainly energy costs to run) ?
I walk you throug a couple of steps to get up and running (covering Windows only) to install a webui to run Stable Diffusion..

Pre-Requirements

• A desktop or notebook with a recent GPU (NVIDIA RTX 30 or 40 series with at least 4GB VRAM)
• Python installed (3.10 recommended) (link)
• git installed (link)

Installation

Open a CMD shell in Windows
Create/navigate to the path where everything shall be installed, at least 15GB space required.

Install the Stable Diffusion web UI (link) using git

git clone https://github.com/AUTOMATIC1111/stable-diffusion-webui.git

Run the webui

webui-user.bat

During the first run Python will create a virtual environment and download/install all packages.

Install the model

Download the Stable Diffusion 2.1 (link) and copy to the models folder

Restart the UI

webui-user.bat

Once the startup script is done, you open the browser at http://127.0.0.1:7860/ and create your first image locally.

You can experiment with the parameters, some are self explaining, others are bit more scientific though.

Errors

You migh get

  File "E:\Python\stablediffusion\stable-diffusion-webui\modules\devices.py", line 150, in test_for_nans
        raise NansException(message)
    modules.devices.NansException: A tensor with all NaNs was produced in Unet.

Run the webui with

webui-user.bat --disable-nan-check

and change the setting

How to design an Airport Self-Service-Kiosk

The Airport Kiosk is a common sight in all small and large airports, it offers a range of self-service features. Upon identification by reading identity cards, credit cards or boarding passes, we can use it to print tickets and boarding passes, print luggage tags, change seat, purchase in-flight-upgrades, spend frequent flyer points and more.
We might reach the stage where these kiosks are a thing of the past, but as long as we need (want) to print something and not 100% of all travellers manage their bookings and ancillary services with their smartphone, we will see them for a some while.

The design and look-and-feel of these kiosks are quite standard, why not create some new designs with our AI wizard assistants ?
I have to admit, I am quite addicted to Stable Diffusion and Midjourney. It is amazing what you can create (or what the AI creates) using nothing but words to describe something.

We are asking for different, fresh designs and want to create an airport experience that is tailored to all kinds of passenger profiles. Let’s give it a try !

We mix the term “self-service kiosk inside an airport terminal” with a variety of art or design styles and settings. Below you find 35 different interpretations. Note, for some art styles, the model didn’t manage to render an airport terminal environment but kept it’s origin. Not all styles are rendered accurately.

Disclaimer: This is solely for educational purpose and does not reflect any actual products or design I am involved with (until today). No copyright infringement is intended.

Technical drawing in the 1970’s

Standard Model

Resembles a mix of ATM and carpark machine.

Family Theme Kiosk

Pikachu

Disney Theme Park

Kids Toy #1 (I quite like this one)

Kids Toy #2

Pixar Style

Phantasy Dreamland Unicorn Style

Kiosk for Elves

Honourable Mention: Kiosk Lego Style

Different eras

Victorian Era

Rococo Era

Gothic Era

Renaissance Era

Romanesque Style

Art Deco Style

1900’s Style

1920’s Style

1930’s Style

1970’s Retro Style (why is it placed in a carpark lot ?)

Steam Punk Style

Mad Max Movie Style

Las Vegas Casino Style

Marvel Theme

Lord of the Rings

Various Designers and Artists

Karim Rashid (Designer)

Naoto Fukasawa (Designer)

Zaha Hadid (Designer)

Roy Lichtenstein (Artist)

Aleksandra Gaca (Designer)

Colani (Designer)

Other Products

Swarovski

Guess what brand …

Google Design

Siemens S62 Rotary Phone Design

Follow this Twitter account for more airport AI art.

Tweets by AirportArtwork

Creating a NFT from Scratch – Airport Terminal NFT

I have been observing the NFT (Non-Fungible-Token) scene for quite a while now. They came into existence in 2012 and get massive attention and hype in 2021 with the record sale of Beeple’s Everydays: the First 5000 Days NFT for $69 million at Christie’s. So far I never mint or bought a NFT, it’s about time to create my first NFT, though a bit of a late adopter for this one.

Disclaimer: This is for educational purpose and not an investment or wealth management advice. NFT’s are highly volatile.

The Basics – What Are NFT’s and How Do They Work?

NFT stands for “non-fungible token,” which is a type of digital asset that represents ownership of a unique item, such as a digital artwork. NFTs use blockchain technology, which is the same technology used in cryptocurrencies like Bitcoin, to ensure that each NFT is unique and cannot be replicated. Digital art refers to art that is created and exists solely in digital form, such as digital paintings, animations, and 3D models. Artists can create and sell digital art as NFTs, allowing buyers to own a unique, one-of-a-kind digital asset. Read more about NFT’s here and the history here.

Artwork Creation

First, let’s create a digital piece of art. Now, with our new little AI wizard friends, Midjourney, Stable Diffusion and Dall-e, that’s a 10 seconds job. But be aware of the imprint of these services, where it is defined if you can use the produced work for commercial purpose, which is clearly the case here, or if you have to apply the Commons Noncommercial 4.0 Attribution International License (or similar). I am paying a subscriber to Midjourney, so this is not problem. Please refer to the terms of services here. Anyway, I do not expect any significant income created with this experiment anytime soon, not yet looking for real estate on Cayman Islands either.

Crypto – You can’t trade NFT’s in cash

We need an account with a cryptocurreny exchange platform, Coinbase is one of the many options. As per legal requirements, you need to verify your identity with them, when you create an account, this is not some dodgy darkweb oeprations.

NFT Marketplace – Sell and Secure with a Blockchain

Our digital piece of artwork does not exist in the physical world, though you could print it on canvas, we have to sell it on a digital platform. There are a few of them, Rarible is my choice.

In order to do any transactions, we need also a cryptocurrency wallet and transfer money into it. We can use MetaMask for this purpose, which is available as browser extension for Firefox.

Transfer ETH to MetaMask

Transaction happened

MetaMask Wallet

You need to link your wallet with the Rarible platform.

NFT Creation – Convert bits of an image file into a NFT

As Simple as selecting ‘Create NFT..‘ and fill out or select some values.

All Done

Et voila ! Our artwork is up for sale ! Link to Rarible NFT.

If you are interested, you can purchase it at $54, it is limited to 10.

Conclusion

• The process is not straightforward and not as simple as uploading an image to a sharing website. You have to have a wallet and know how cryptocurrencies work. Maybe nothing that my mom (Sorry!) could manage to do.
• I spent more time writing this post than the actual process itself. Is this art ?
• Part of above text was even created with ChatGPT !
• IMO it seems the hype is wearing off and the whole Metaverse and NFT hype starts to fade slowly ?

https://www.theblock.co/data/nft-non-fungible-tokens/nft-overview/nft-trade-volume-by-chain