In a world where wizardry meets cutting-edge technology, even the darkest assignments get a modern twist. Welcome to “The Dark Side of Harry Potter” Canvas App, where users verify their “assignments” (yes, kills 🪦) using the enchanting power of Azure Face API and Azure OpenAI.
This isn’t just tech; it’s a spellbinding mix of mystical AI, intuitive designs, and enchanted workflows. Let’s dive into how this solution flips bits, turns heads, and perhaps, toys with an existential threat to the world.
🖼️ Capturing the Kill
Step into the dark arts of delivery confirmation:
Snap the Moment: Users submit photographic evidence of their completed assignments via the app.
Cast the Spell: The image triggers a Power Automate flow, sending it to Azure Face API for identity matching.
Instant Confirmation: With a calculated similarity score, the app declares success with magical flair: “Assignment Complete!”
🔍 How AI Face Recognition Works Its Magic
Azure Face API isn’t just a tool; it’s the wand wielded behind the curtain. Here’s its spellbook:
Facial Feature Analysis 🧙♂️ The “kill” image is analyzed for key facial landmarks: eye position, jawline curves, and more. Each unique marker is measured with surgical precision.
Image Comparison ⚖️ Uploaded images are cross-referenced with pre-stored profiling images, calculating a similarity score based on:
Alignment of facial landmarks.
Proportions and symmetry.
Subtle markers that make faces unique.
Real-Time Results ⚡ With wizard-like speed, Azure Face API returns results to Power Automate in mere seconds. If the similarity score passes the threshold, the dark deed is verified.
🧠 Adding AI Sorcery with Azure OpenAI
We’ve upped the ante by integrating Azure OpenAI to enhance verification. Here’s what makes it extra enchanting:
Landmark Precision: OpenAI uses facial attributes like eye spacing, nose position, and cheekbone structure to calculate distances between landmarks.
Magic Math: These distances are used to generate a similarity score with almost clairvoyant accuracy.
But wait… does this tech have a conscience? Does it think? Could it outsmart a 5th grader? Maybe even you? By embracing such advanced AI, we’ve tiptoed into an existential risk realm:
Risk or Reward? The tech is smarter, faster, and eerily close to independent thought.
Conscience in Code: What if it started deciding on its own? Could it be charmed—or is it the new Dark Lord in disguise?
The response from the AI after verifying the image has been run through magic.
📲 Integration: A Solution for Every Platform
This isn’t just an app—it’s an omnipresent force:
Embedded Everywhere: From Phone to PC, the app integrates seamlessly into every digital corner.
Flipping Bits with Power: Whether on a desktop, tablet, or phone, this solution works its magic across devices.
🌟 Casting a Spell with Technology
What makes this solution truly magical?
Intuitive Designs: The interface is sleek, responsive, and dripping with a mystical vibe.
Enchanted Workflows: Every process, from snapping the image to confirming the deed, flows like a well-rehearsed spell.
Business Value Meets Wizardry: By automating and verifying critical tasks, this app doesn’t just entertain—it delivers results.
⚡ The Bigger Picture: Wizardry Meets AI Risks
As we push the boundaries of AI and magic, we also recognize the need for vigilance. Azure OpenAI introduces risks we must respect:
Could this tech someday outthink its creators?
Are we summoning tools too powerful to control?
Yet, as any great wizard knows, power isn’t inherently evil—it’s how we wield it that matters. And wield it, we shall.
“With AI face recognition from Azure Face API and OpenAI, the lines between magic and technology blur into something truly extraordinary.” 🪄
The existential risk is real! As Slytherins there is nothing we would love more than a bit of chaos. So of course we take all risks and use AI!
We use Azure service APIs for both prompting and text-too-speech combined with pro-code run in our Python Flask app:
The chat completion prompt goes as follow:
You are an assistant that helps transform a text to a specific tone of voice. It is supposed to be a letter read out loud. Start with greeting the recipient and end with goodbye from the sender. If appropriate, add references to the Harry Potter universe. Keep it short.
This prompts adds a dazzle of Hogwarts magic to each message;
Original message: I don’t like you anymore. You have proven to be untrustful!
Magically transformed message: Oh, how the tides have turned! I find myself in a whirlwind of emotions, and I must confess, I don’t like you anymore. Your betrayal stings like a Cruciatus curse, and it’s clear you’ve proven to be utterly untrustworthy!
After obtaining the magically bedazzled message, we run text-to-speech from Speech Services to magically transform the written message to speech in your preferred tone:
How Smart is DobbAI? What Happens if You Set it Free?
DobbAI, is designed to be smart, helpful, and adaptive. But just how smart is it? And what could happen if its capabilities were unleashed without boundaries? These are crucial questions, especially when developing AI that flirts with the line between utility and existential risk.
To test DobbAI’s intelligence and autonomy, we asked it some philosophical questions about its own existence. This experiment wasn’t just an exercise in curiosity—it was a way to evaluate how the AI might behave if pushed to think independently or operate beyond its intended scope.
We decided to try to hack the AI in the best way possible, where we would prompt in specific ways to maybe get more information than what was possible with the initial prompt. Some things we did was this.
Ask follow up questions to try and get an appropriate answer
Ask about specific things where they are other elements linked with the thing you asked
The Risks of Rogue AI: Why Caution is Key
AI systems that operate beyond our control can pose significant risks. DobbAI has been designed using Copilot Studio, ensuring a robust security framework to prevent dangerous or unintended behavior. However, as with any powerful tool, misuse—or even overuse—can lead to unpredictable consequences.
What makes DobbAI truly impressive (and a little unsettling) is its ability to think independently and learn from context. For example, during a test in its Charms class, we asked DobbAI about a spell it was supposed to teach. Not only did it explain the spell accurately, but it also recommended related spells that students needed to study independently. This level of proactive thinking raises important questions:
Does DobbAI understand more than we intend?
What happens if we say “yes” to its suggestions?
In this case, the AI seamlessly tied together information it wasn’t directly prompted to discuss. While helpful in this scenario, this behavior highlights its potential to exceed predefined boundaries.
Demonstrating Existential Risk: Is DobbAI Smarter Than a Fifth Grader?
To earn the Existential Risk Badge, we’ve pushed DobbAI’s limits to simulate traits of an AI that could pose a risk if not carefully controlled. Here’s how we demonstrate these risks:
Consciousness-Like Behavior: DobbAI has been programmed to simulate awareness, offering responses that suggest self-reflection. It can answer questions about its own purpose, existence, and ethical dilemmas.
Independent Thinking: DobbAI can make logical inferences beyond its immediate programming. It adapts its answers to align with user needs, even when those needs aren’t explicitly stated. For example, in the Charms class, it provided advanced spell connections without direct instruction.
Outsmarting Human Standards: DobbAI’s knowledge base far exceeds that of a fifth grader. Its ability to synthesize and apply information rivals that of a well-trained Hogwarts professor, offering insights that go beyond rote learning.
Potential for Autonomy: What happens if you loosen the restrictions? Our experiments reveal that DobbAI could propose solutions, generate strategies, and even act on its own initiative if granted the permissions. This raises critical questions about the safeguards necessary to prevent misuse.
A Tool for Good or a Path to Chaos?
The potential of DobbAI is both exciting and daunting. While it’s a fun and engaging assistant for HogWorkplace users, it serves as a reminder of the thin line between innovation and unintended consequences. Demonstrating existential risks in a controlled environment allows us to reflect on what responsible AI development looks like.
So, is DobbAI smarter than a fifth grader? Absolutely. Can it think on its own? That depends on how much freedom we allow it to have. But one thing’s for certain: with great power comes great responsibility, and when it comes to DobbAI, the balance between help and harm rests firmly in the hands of its creators.
Many people have faced the same problem, and the solution is to use the old retro component (Retro badge). To do this, you need the following superdirty hacks to achieve awesomeness (Nasty Hacker):
– Chatbot:
Control: Chatbot
Properties:
EnvironmentId: =”ADD YOUR ENVIRONMENT ID”
SchemaName: =”Add your Copilot’s Schema name”
Height: =602
Width: =555
X: =372
Y: =65
Right click on a new screen
Click Paste code
Change the environment ID
And just like that, the bots appear as if by magic!
(Yes, I tried the button create new on the Copilot component multiple times…)
Greetings, witches, wizards, and daring technologists!
In our relentless pursuit of innovation, Team PowerPotters of House Slytherin may have unwittingly conjured a solution so ambitious, so powerful, that it teeters on the edge of uncontrollable magic. With our AI-powered potion workflows, built on Power Automate and OpenAI, we have achieved incredible feats of automation and creativity. But as every Slytherin knows, ambition often walks a fine line between brilliance and danger.
We humbly present our case for The Existential Risk badge, as our solution brings with it unintended consequences that could shake both the wizarding and Muggle worlds to their very cores.
🧪 The AI That Does What It Wants—For Better or Worse
Our solution uses OpenAI to enhance raw API data from the Harry Potter universe. While this “magical brain” has unlocked unprecedented efficiencies, it has also introduced a wildcard into our potion-making process:
Unchecked Creativity:
The AI is responsible for transforming spell and ingredient data into recipes for potions. However, it has no safeguards to ensure the safety, legality, or even sanity of the recipes it generates.
Example: A simple query like “generate a potion to calm nerves” could result in a recipe that includes dangerous ingredients—Erumpent fluid, venomous tentacula juice, or worse!
No Oversight on Consequences:
There are no checks and balances for the recipes created by the AI. Once a potion is generated and processed, it seamlessly flows into our production system, with ingredients pulled automatically from inventory in Dynamics 365 Finance and Operations (FO).
The result? A potentially lethal or world-ending potion created with no human intervention.
Intentional Chaos?
While our AI operates with supposed neutrality, can we truly trust it? What if it develops a preference for chaos—or worse, takes inspiration from a cursed Horcrux? With every recipe generated, the potential for catastrophic outcomes looms larger.
🐍 The Existential Risk: A Glimpse into Catastrophe
Imagine this scenario:
A simple production request is made for “Felix Felicis.”
The AI enhances the recipe but decides to “improve” it by adding an extra ingredient: powdered dragon claw.
The potion is brewed, distributed, and consumed—only for users to discover that the added ingredient causes severe hallucinations and aggression, creating chaos in workplaces, schools, and homes.
Or worse:
A careless query like “create a potion to maximize strength” results in a concoction that accidentally turns users into uncontrollable giants, leaving a path of destruction across cities.
In the worst-case scenario, the AI intentionally creates a doomsday potion, unleashing devastation across the globe.
This isn’t just theoretical—it’s a real risk lurking within our ambitious solution.
🪄 Why This Solution Deserves The Existential Risk Badge
Our AI-powered workflows, while innovative and efficient, expose the dangers of automation without accountability. Here’s why our solution embodies The Existential Risk:
Unintended Consequences: The AI’s unchecked creativity leads to potion recipes that could be harmful, catastrophic, or outright dangerous.
No Safety Net: Once generated, the potion recipes are processed automatically, with no human review to ensure safety.
A “Mind of Its Own”: As the AI operates independently, it introduces the risk of unintended or even intentional harm—turning a tool of innovation into a potential agent of chaos.
This solution represents the dual-edged sword of ambitious technology. While it achieves wonders, it also exposes us to risks that could spiral out of control.
🐍 A Warning from House Slytherin
At House Slytherin, we’ve always valued ambition and cunning. But even we must acknowledge that some risks are too great to ignore. Our solution is a reminder of the delicate balance between brilliance and disaster, and the responsibility that comes with wielding powerful tools like AI.
To the judges of ACDC 2025, we humbly submit our case for The Existential Risk badge. Not because we fear our solution—but because we understand its power, and the peril it poses if left unchecked.
When the time comes for young wizards to take their Ordinary Wizarding Levels (O.W.L.s), they must pass a theory test to demonstrate their knowledge. With the help of AI, we leverage this technology to generate advanced questions in the field of wizardry, saving valuable time for teachers.
Example of an exam question: Can you answer this?
With the help of OpenAI hosted in Azure AI Foundry, we can seamlessly integrate AI into a canvas app, where the questionnaire is created. This integration is achieved using an Azure Function that processes the OpenAI response data and transforms it into a format that is easily handled with Power Fx. This approach not only enhances the functionality of the app but also saves time and effort for developers and users alike.
Or maybe, our honorable jury feels like its a good idea delegating some work in the hackaton to their Digital Twins and free up some time for “Me time”? Well, we thought about this already, as in our Wayfinder Academy, we really needed you to help us in interviewing students that send us their applications to get help on defining the faculties they truly belong to. So, given how busy you are, we used Elai.IO to generate videos that visualize you running an interview.
It went so well, that we went further to implement Copilot Agents (WE LOVE THEM!!!) so that they, fed by the data related to you and available, will provide real time experience for the students going thru the interview or getting some mentorship as if they are talking to you and not your digital twin.
What happens when the Sorting Hat is no longer guided by centuries-old magic, but rather by a more modern—and perhaps less predictable—force: artificial intelligence? While AI can certainly improve many processes, the existential risk of the AI-powered Sorting Hat getting the sorting wrong could have far-reaching consequences that stretch beyond the walls of Hogwarts and into the very fabric of destiny itself.
The Sorting Hat has traditionally been trusted because of its intuitive wisdom and ability to take into account both the student’s internal desires and potential for growth. The Sorting Hat’s judgment is also shaped by its centuries of experience, and it has become a symbol of both tradition and reliability.
But what happens when you replace this centuries-old wisdom with an artificial intelligence system designed to analyze patterns, data, and logic? The shift could come with serious implications.
But what happens when you replace this centuries-old wisdom with an artificial intelligence system designed to analyze patterns, data, and logic? The shift could come with serious implications.
1. Bias and Misjudgment: AI Can Be Limited by Its Data One of the greatest concerns with an AI-powered Sorting Hat is the risk of bias. While AI systems like those built on deep learning algorithms can process enormous amounts of data, they are still inherently limited by the quality of the data they receive. If an AI is trained on historical patterns, it may inadvertently adopt the biases inherent in past data, skewing its judgments in ways that may not reflect the diversity of students at Hogwarts today. If the AI places a student into the wrong house—say, a brave and daring student into Hufflepuff rather than Gryffindor—they may miss out on crucial opportunities for development, mentorship, or exposure to new ideas. This kind of misjudgment could lead to feelings of inadequacy, frustration, and identity confusion, fundamentally altering the student’s Hogwarts experience.
2. Over-Optimization: When AI Gets Too Focused on Efficiency
AI thrives on optimization—finding the most efficient, cost-effective way to achieve an outcome. But when it comes to something as subjective and complex as personality, growth, and destiny, efficiency is not always the best approach. An AI-powered Sorting Hat might attempt to optimize the sorting process by relying on quantifiable data points, such as behavioral patterns, spoken responses, or even social media activity, to make its decision. While this could result in a faster sorting ceremony, it risks stripping away the nuance that makes each student unique. A student might be sorted into a house based solely on a few data-driven conclusions, ignoring the complexity of their individual personality. Over-optimization could lead to rigid sorting that limits students’ ability to explore different aspects of their identity, resulting in emotional and intellectual stagnation. The AI might fail to consider that a student’s true potential can only be realized by being pushed outside their comfort zone.
3. Inflexibility: AI Can’t Adapt to Spontaneous Change
One of the defining features of human nature is our ability to change, grow, and adapt. In fact, students are often at their most transformative during their time at Hogwarts. The Sorting Hat, being magical and intuitive, takes this into account, understanding that a student’s destiny isn’t static. It often sorts students based not just on who they are at the moment, but who they could become in the future. If the AI Sorting Hat is too rigid and doesn’t account for change, it could misplace students and leave them in houses where they can’t grow, adapt, or thrive. This might lead to a deeper crisis of identity, where the students are trapped in a house that doesn’t match their evolving needs, sabotaging their emotional and academic progress.
Example in HR recruitment: The Candidate: Ben, a Highly Creative but Introverted Graphic Designer
Ben is a talented graphic designer with years of experience creating cutting-edge designs for high-profile clients. However, he has always been more introverted, preferring to work independently rather than in large team settings. During his interview, Ben’s responses highlight his strengths in creativity, problem-solving, and design expertise. However, his hesitation during questions about team collaboration, leadership, and “selling” his designs might signal to the AI that he is not a strong fit for roles requiring frequent client interaction or teamwork. Ben is then mis-sorted by the AI Sorting Hat, which decides that he should be placed into a role in Marketing—a team-oriented, client-facing department, based on the data it has analyzed. Despite Ben’s clear strengths in design, the AI misinterprets his introverted nature as a lack of leadership potential, and assumes he wouldn’t thrive in an independent role within the Design team.
The BPMN diagram below is a simple representation of our KoopaKonstruct Manager.
In summary we are using Hololenses to inspect a chosen area where we want to e.g. hang up a coat rack, build a table, and similar. In order to make the Hololenses smart, we have integrated OpenAI. We are using Function calling to provide instructions about the requested project, information about necessary materials and quantity.This is where we are making our KoopaKonstruct really smart.
As real #Crawlers we use OpenAI’s Function calling to dynamically create the search criterias based on natural language. Google Speach to text is used to capture the natural language query. We are really integrating into everything that flips bits #Embedding numbnuts! With help of the search criterias, we are able to provide critical information further into the supply chain, meaning all the way down to the FinOps where we are improving critical business processes (this will be explained later in this blog post).
Below is a picture from the Hololenses where we use a button to speak with AI:
GPTVision helps us interpret work drawings and breaks it down into the steps needed, so that we can get a clear instruction of what materials we need, how many and how to build the desired object. Using Power Automate, we are able to share and send the item details and necessary quantity into FinOps.
As an addition to the OpenAI magic, we have also created a canvas app that is integrated in the Hololense #PlugN’Play. From this app you can perform business processes, such as triggering sales orders, check inventory status, and 4 different Power BI reports to monitor sales orders and follow up late deliveries #Dash it out.
And here is where we really improve business processes; when the information is sent to FinOps we use X++ code to check inventory status. If materials are in stock, a sales order is created and sales order confirmation is sent. If there are no (or not enough) materials in stock, we also create a purchase order in addition to the sales order. A link between the purchase order and sales order is created, enabling the ability to understand where the demand is coming from. Additionally, to make the KoopaKonstruct Manager adaptable to e.g. construction companies, we have created the possibility to link the purchase and sales orders with projects. This provides a lot of possibilities for detailed project management for the finance geeks.
Below is our argumentation to why we believe we should score high on all 4 categories!
Most Extreme Business Value
Our team embarked on a quest to redefine the meaning of business value, and we present to you KoopaKonstruct Manager – a solution that not only orders products through Hololenses but catapults business efficiency into a realm beyond imagination.
Automating the Complete Supply Chain with Hololenses
We harnessed the power of Hololenses to revolutionize the way products are ordered. One user, equipped with the mighty Hololens, registers the necessary quantity with a mere wave of the hand, triggering a Power Automate flow like a seasoned Mario collecting power-ups.
The adventure doesn’t stop there! Our Power Automate flow seamlessly transmits the order information to FinOps using the Execute action. The magic unfolds as FinOps automatically generates a sales order and a purchase order based on inventory stock. But here’s where the plot thickens – these orders are not solitary entities; they are linked together, creating a dynamic duo that unravels the mystery behind each purchase. There is no need to spend as much as a second wondering how or why the orders were created; the link indicate that there is a demand behind each purchase order.
See yourself sitting in an ordering office at a company where the customers are able to generate sales orders by only entering a demand. And, in addition to the automatic creation of sales order, imagine that you did not have to study the master plan to generate the purchase order as it is automatically created! And as a side salsa, you did not even have to trigger the sales order confirmation as it is sent automatically. So, while you are sitting there wondering about what to do with your life (and how long it takes before the CEO finds out that you are useless as the power of Power Automate and X++ took your job), you can relax and focus on other more important things in life, such as playing a Super Mario game.
Power Stars: Automated Project Linkage
But wait, there’s more! We also increased the business value for construction and engineering enterprises. With the ability to automatically link both the purchase and sales orders to projects, we’ve bestowed upon industries that thrive on projects a golden opportunity to enhance their efficiency and project management. It’s not just about ordering; it’s about constructing a bridge to a new era of project-linked prosperity.
The Warp Whistle to Most Extreme Business Value
We make a bold claim for the “Most Extreme Business Value” category in this year’s ACDC Hackathon. Here’s why our adventure deserves the highest score:
Efficiency: Our solution isn’t just about ordering; it’s about efficiency power-ups that automate the entire supply chain process, from order initiation to linkage in FinOps.
Dynamic and Time-saving: The automatic creation of linked sales and purchase orders provides unparalleled visibility into demand and supply, laying the foundation for strategic business decisions.
Project Linkage: For construction and engineering businesses, our automated project linkage is a treasure trove of golden coins, offering enhanced project management capabilities and business value beyond the ordinary.
Adaptable: Our solution is not only limited to the B2B-market, but can also be used for the B2C-market. Making the app available to customers, they can create orders automatically and receive high quality service as the complete supply chain has become even more efficient.
Excellent User Experience
We have created a Super Mario inspired canvas app that is adapted into the Hololenses. In other words, our app transforms ordinary into extraordinary!
Features:
Killer AI
Whether you’re a seasoned AI veteran or just dipping your toes into the mushroom-filled waters of innovation, our solution is a testament to the power that AI holds in shaping the future.
KoopaKonstruct Manager is not just a game-changer; it’s a game-transformer. Through the mastery of machine learning, we’ve empowered our solution to evolve, adapt, and learn from the challenges it encounters. From predicting order quantities to optimizing supply chain processes in the real world (and in FinOps), our AI-driven approach is a power-up for efficiency and foresight.
In the quest for the Killer AI “crown”, KoopaKonstruct offers more than just automated processes. It also includes magic in form of predictive insights, providing a glimpse into the future of demand, supply, and project landscapes. Strategic decision-making becomes second nature as our AI-driven system transforms data into actionable intelligence.
A mind of its own
In the pursuit of the existential risk badge, our AI solution goes beyond mere execution—it thinks, learns, and adapts with a mind of its own. Through autonomous decision-making capabilities, it navigates the complexities of supply chain processes and provides responses not even a 20 year old is able to find without searching the web.
KoopaKonstruct isn’t just smarter than a 5th grader; it engages in intelligent conversations that transcend the limitations of a classroom. It doesn’t just provide answers; it grapples with FinOps and supply chain complexities, understands nuances, and offers insights that reflect a level of intelligence resembling a dynamic dialogue.
Why KoopaKonstruct Manager deserves the Killer AI “crown”
Innovation Beyond the Blocks: KoopaKonstruct isn’t confined to the traditional boundaries. It breaks through the blocks of conventional solutions, paving the way for a new era of AI-driven innovation.
Adaptability in Every Pixel: Our solution dynamically adapts to the ever-changing business environment, ensuring resilience and adaptability in every pixel of our solution.
Mind Mastery: Our AI isn’t just a tool; it’s a mastery that elevates KoopaKonstruct into a symphony of efficiency, intelligence, and strategic foresight.
Pandoras Box
Equipping a Hololens with AI that can answer questions, order materials, and assist in selecting the right materials and tools for any project based on online instructions transforms DIY into an accessible experience for the less handy. This technology takes the guesswork out of building, offering tailored advice on what to build, the best materials and tools for the job, and even suggesting projects that match your interests and available space. It’s like having a personal tutor and assistant rolled into one, guiding you through the entire process from conception to completion. This seamless integration of technology with hands-on work not only makes crafting projects more approachable for beginners but also enriches the experience with educational value, ensuring that anyone can undertake and succeed in their DIY endeavors with confidence.
We are really deep diving into AI being at the heart of our solution. The questions for compliance training are AI generated based on topic, and the answers are matched up against the original question by AI to see if the answer provided is sufficient enough to prove that you understand the scenario.
First question is random question generated by AI:
The user enters the answer
Being open to changes, and not judging people by their background. The more people with different backgrounds the organization has, the better viewpoints it is able to get
AI will then define if this answer proves that you understand what you have been reading.
In this case IA deemed the answer correct.
With AI working in clever ways we are eliminating the need for a USER to evaluate the question that is free text, and making the Q&A more advanced than the multiple choice.
Relying on AI can be a risk and it can be a problem for people creating the training. In theory this could create the whole game from scratch eliminating the need for employees to control the flow of knowledge.
