As a lifelong Shark Week enthusiast with a budding interest in surfing, I've always found myself in awe of the ocean and its inhabitants.
Growing up in Connecticut, my journeys to the ocean were often limited to the summer, with each trip feeding my growing fascination with marine life.
A couple of years ago, I decided to turn my fascination into a hands-on hobby.
I took up surfing.
Starting, I was pretty terrible. I still am, to be honest, but I've loved every minute I've spent in the waters of Narragansett, Rhode Island.
Simultaneously, I was diving into another passion of mine - artificial intelligence (AI).
Back in college, I was a computer science minor, but my interest skewed toward the data science and AI components.
There was something magical about code and algorithms turning raw data into meaningful patterns and predictions.
During one of those sweltering summer days when I was neck-deep in an AI exercise, the idea for SafeWaters.AI hit me.
The task at hand was developing AI models to predict crime rates in specific areas and the chances of developing cancer-based on lifestyle habits.
Later that day, a news headline caught my eye.
Eight great whites had been tagged in Narragansett, where I'd been trying to surf just days before.
AI to Forecast Shark Crime
Suddenly, it clicked.
What if I could apply these AI algorithms to forecast what I dubbed at the time as "shark crime"?
The concept of using technology to make our interaction with the ocean safer was captivating.
I was on fire with this new project, sketching out plans and looking at domains.
Initially, I aimed for "shark.ai," but it was already taken. That's when "SafeWaters.AI" popped into my mind, and it just felt right.
After all, the ultimate goal was to ensure safer waters for everyone.
At first, the SafeWaters algorithm was being built on the side.
I was a full-time freelance SEO specialist, so SafeWaters was my after-hours project.
It took several months to locate, gather, combine, and organize the data needed to train such a complex problem.
This was the biggest challenge. I also, by no means, was an AI professional.
Instead, I was building this project as a self-taught in the art of machine learning and artificial intelligence.
Luckily, this amazing assistant I hired named ChatGPT was great at debugging code and making recommendations.
I even had this assistant create a course for me to learn everything I needed to complete this specific model.
The Winning Model
After several different data sets, janky models only forecasting with 11 percent accuracy, some even 100 percent (which I debunked as having a data leakage problem), I built the final winner, training on 200 plus years of attack and marine weather data, forecasting our test data with an 89 percent accuracy.
No issues, no bugs, no data leaks.
Today, I'm living my dream in Boston, Massachusetts, solopreneuring SafeWaters.
It's an exhilarating journey of harnessing the power of AI to predict the risk of shark encounters with striking accuracy.
SafeWaters.AI is much more than a static app. Our AI model constantly learns and evolves, improving its predictions over time.
We've meticulously fed it a diet of historical shark attack data and marine weather conditions.
Its accuracy has been tested on over 1,600 previous attacks, with the model correctly categorizing 89 percent of those days as "high-risk."
It's not just about creating an app; it's about equipping individuals with potentially lifesaving information.
It's important to emphasize we are not "predicting" a shark attack but forecasting the risk of one.
A high-risk day is when all the variables seem to line up. We don't want to drive chaos at the beach.
We just want to encourage informed decisions and precautions like maybe taking off the shiny jewelry, staying shallow, or hiring extra shark spotters if your beach takes advantage of them.
Funding the Running Costs
SafeWaters.AI isn't just about human safety; we also care about our finned friends.
By reducing unintended encounters, we hope to reduce harm to these magnificent creatures.
We're committed to giving back, pledging to donate 5 percent of our profits to ocean cleanup initiatives.
This project isn't just a testament to technology's potential; it's also a love letter to the oceans and their inhabitants.
Right now, we've got a Kickstarter campaign underway.
It's not just about funding; it's about sharing our journey and inviting you to be a part of it.
We're offering early bird discounts to those who believe in our mission and wish to support us.
This will help SafeWaters with the costs of hiring the best high-class mobile app developers and cover additional running costs.
As I said, I really focused on the data science side of computer science, so these mobile developers are crucial for the launch of SafeWaters this summer.
I know it may seem like we're asking for a lot, but in the grand scheme of things, it's all about making our shared oceanic space safer for everyone.
And trust me, this is not an attempt to sell you something; it's more like an invitation to join our cause.
You can check out our Kickstarter here if you're interested: https://bit.ly/safewaters-ai
A Lifesaving App
In this journey of developing SafeWaters.AI, one thing has been made abundantly clear to me.
AI isn't just about creating flashy robots or self-driving cars.
It's about using technology to enhance our lives and, in our case, making the ocean a safer place.
When I think back to the days of watching Shark Week with wide-eyed fascination, I never imagined that I would one day be working on a project like this, harnessing the power of AI to not only protect us humans but also these majestic creatures of the sea.
To be frank, the journey hasn't been all smooth sailing.
There have been countless late-night coding sessions, periods of doubt, and bouts of frustration when things didn't work out as planned.
But every challenge and every setback has only reinforced my belief in what we're doing.
Every time we make a breakthrough, every time our model successfully predicts a high-risk day, I'm reminded of why we started this in the first place.
In the grand scheme of things, SafeWaters.AI is just a drop in the ocean of AI's vast potential.
Its implementation in different facets of our lives, from healthcare to transportation to environmental conservation, is reshaping our world.
And I'm beyond stoked to be a part of this wave of innovation.
What started as a fascination with sharks, a love for the ocean, and an interest in AI has evolved into something much more meaningful.
Every line of code, every test run, and every update takes us one step closer to our goal of safer waters for everyone.
And as much as it's a testament to the power of AI, it's also a testament to the power of a simple idea born out of passion and curiosity.
SafeWaters.AI is more than just an app; it's a mission, a vision, a dream fueled by passion, hard work, and a little bit of sea salt.
It's about creating a symbiotic relationship between humans and the ocean, using technology to foster understanding, respect, and safety.
It's about riding the wave of AI innovation, not for the thrill of it, but to make a difference.
As we look towards the future, the vision for SafeWaters.AI remains clear. Much like our AI, we strive to continue improving, evolving, and learning.
The ocean is full of mysteries and challenges, but with every stride, we're making it a little bit safer for everyone.
And in the end, that's what SafeWaters.AI is all about - a safer, cleaner, shark-friendly ocean.
So, as we embark on this journey, I invite you to join us.
Let's ride this wave together toward safer waters, a greater understanding of our marine life, and a future where every week could be like Shark Week - not filled with fear, but with understanding and respect.
Words by Evan Valenti | Founder of SafeWaters.AI