Whilst AI has transformed many essential areas of society, some have also begun to use its technology for the weirder. As artificial intelligence becomes more and more a part of everyday life, people are being more creative in implementing its technology into a variety of different areas, from the mundane to the wonderfully mad, often with comical results.
Without further ado, here is our list of the top 5 strangest uses of AI:
Finding Waldo
The creative marketing agency Redpepper have developed a robot to help find Waldo in the famous book series Where’s Waldo (known as Wally in the UK). Researchers used Google AutoML Vision, feeding in images of the striped, elusive character to help identify the striped character in the puzzle books’ crowds.
When presented with one of the Waldo/Wally pages, the robot (an arm with a camera on the end and tiny plastic hand) will take a picture of the scene, compare it to the images it’s been fed, find the spectacled traveller and smack the plastic hand down onto him.
Writing Pickup Lines
Janelle Shane, a research scientist, trained a neural network in how to create pickup lines. For this, Janelle fed pre-existing pickup phrases to the neural network, which it then analysed in order to help generate lines of its own. The results, however, were not as silver tongued as one would expect, but rather oddly endearing, if slightly incoherent, including:
“You look like a thing and I love you.”
“I want to see you to my heart.”
“You must be a triangle? Cause you’re the only thing here.”
Flipping Burgers
Miso Robotics have created a robot that can flip burgers (aptly named “Flippy”). It’s claimed that the artificial cook can grill up to 300 patties an hour, as well as offering a whole host of other culinary skills, including:
- Working a grill or fryer
- Automatic tool switching and cleaning
- OSHA safety compliant
- Works collaboratively with kitchen staff
Flippy also cooks “perfectly and consistently everytime” and can “learn from its surroundings and acquire new skills over time”.
Fortune Telling
Roboticist Alexander Reben has created a virtual fortune cookie, taking thousands of the fun treat’s one-line predictions and training a neural network on them. The results for Alexander’s experiment was quite unsettling, with around 75% of the messages being negative, and some just plain odd, including “Success is a powerful excuse” and “no one is listening.”
Writing Harry Potter Books
Whilst J.K. Rowling saddened many fans by bringing the Harry Potter series to a close, AI tech has made it possible to create more of these magical books without the author. Botnik studios wrote a new chapter to the Harry Potter books, using an algorithm that has been trained on the seven existing Potter books.
The new, four-page chapter Botnik studios created was titled “Harry Potter and the Portrait of What Looked Like a Large Pile of Ash” and holds comedic, if slightly confusing, situations such as - “Voldemort, you’re a very bad and mean wizard,” Harry savagely said. Hermoine nodded encouragingly. The tall Death Eater was wearing a shirt that said ‘Hermoine Has Forgotten How To Dance,’ so Hermoine dipped his face in mud.”
What are the four types of artificial intelligence?
Reactive machines, theory of mind, limited memory, and self-awareness are four main types of AI categories to be aware of. Read more about them in this guide.
What are the 3 V's of big data?
What are the 3 v’s of big data? Find out more as the data experts at Optalitix explain what they mean by using examples to illustrate each of them.
How to Become a Data Scientist
When pricing a new product, the starting point is almost always Excel. Learn how to create a pricing application from a spreadsheet pricing model here.
Holiday AI Thoughts
The key to selling is to establish a relevant dialogue and understanding with your clients. Find out how AI systems can be used to target the right consumers.
Artificial Intelligence powered insurance is now inevitable!
AI is changing the competitive landscape for insurers by delivering a competitive edge in a number of areas. Read more about AI in insurance.
AI Accessible for all - Ignore at your peril!
As AI continues to develop, businesses of all sizes will need to engage with the technology and embrace the power of machine learning to improve their processes.
The Computable Game
Data analysis with Optalitix—an online underwriting tool offering advanced data-based services, including database management and analytics to insurers & bankers.
AI and machine learning in a GDPR environment.
GDPR refers to the EU’s upcoming data protection regime. Find out whether the anonymisation of data, as prescribed by GDPR, would break AI models and more here.
Cultural diversity drives better AI outcomes
Agile organisations that can respond to the global challenges of their often multinational customer base are often the ones that thrive. Read more in this guide.
Why cant lending be like Uber?
Lending is constantly changing with entrants, regulations and technology. Learn how technology can give businesses market leading edge like Uber in this guide.
Why are Insurers playing a guessing game?
Numerous insurers overlook advanced solutions. Make sure your company leverages valuable insights from its data to enhance the user experience with Optalitix!
The future of life insurance
In the past, choosing the right life insurance was an arduous task. Thanks to data analysis and algorithms, shopping for life insurance has become much simpler.
Insurance quote personalisation is magic!
Dani Katz talks about how Optalitix can improve every aspect of the insurance business through better use of data, quotations and more. Read more in this guide.
A meaningful connection. The next phase of API relationships.
Data analytics and API closely intertwine, but what is the next phase in their relationship? Learn more about this successful financial connection here.