Why the rise in AI sports betting predictions comes with serious risks
AI is becoming a bigger part of the sports betting world, and more and more players appear to be turning to bots to help them make predictions, even though experts warn that serious risks come with it.
Searches for terms like “AI sports betting predictions” have exploded recently, jumping anywhere from 200% to a staggering 5000% in just the last month, according to Google Trends.
The surge might have something to do with influencers talking up their results and a growing sense of trust in AI’s potential. The industry could even balloon to $29.7 billion by 2032.
Interest in AI-driven sports betting is surging worldwide. Credit: ReadWrite / Google
However, while many betting apps use their own built-in AI tools, some people are leaning on platforms like ChatGPT or Gemini. The problem is that these external tools can come with major pitfalls, such as AI generating fake stats that sound convincing or giving out unreliable financial advice. So while the tech is exciting, it definitely comes with a “proceed with caution” label.
When influencers share videos of their so-called ‘safe picks’ and confidently show off their favorite AI prompts, it can make sports betting look like an easy way to make money.
But the reality is that AI tools like ChatGPT do not actually understand betting odds, market dynamics, or insider stats. They are also known for repeating false or outdated information, as we’ve previously reported. The tools make predictions based on whatever they find online, and that information is not always reliable.
Well-known betting sites usually do a good job of explaining the odds and risks. But if you’re using something like ChatGPT and just going with whatever it says without thinking it through, that can be misleading. The AI might sound confident, like it knows what it’s talking about, but it doesn’t actually understand how betting odds work.
How AI is transforming sports bettingThe amount of money being poured into AI shows just how seriously bookmakers are taking this technology. According to Allied Market Research, in 2022, the global market for AI in sports betting was worth about $2.2 billion.
Analysts believe it’s only going to grow from here, with projections showing a compound annual growth rate of 30.1% between 2023 and 2032. By then, spending on AI sports betting strategies could hit $29.7 billion.
AI is transforming the sports industry by enhancing performance, fan engagement, and business operations. In 2024, AI's market impact is expanding rapidly, driven by its ability to analyze vast amounts of data to optimize player performance, prevent injuries, and develop game… https://t.co/nWTzCvuHIc
— Allied Market Research (@allied_market) August 8, 2024
It has been said that a rising demand for things like player data tracking, chatbots, and real-time insights is the reason behind this growth. Bookmakers are racing to roll out these features and stay ahead of the competition.
In the past, sports betting relied on static models for analysis but the models couldn’t adapt on the fly or handle live data. On the flip side, machine learning uses algorithms to process real-time data and identify patterns that can lead to smarter predictions. This includes tracking player performance trends, analyzing past scores, and even spotting shifts in the betting market as they happen.
Is there an AI that can predict sports betting?There are AI systems that can help predict sports betting outcomes, but it’s important to understand how they work and what they can and can’t do.
FanDuelThe systems look at all kinds of factors, such as player stats, injuries, weather conditions, and even live game feeds, to help set odds that reflect the current situation.
For example, FanDuel uses the tech to power its NFL play-by-play betting. As a result, they can adjust odds on the fly as the game unfolds, creating a more dynamic and responsive betting experience.
DraftKingsAnother area where AI is making an impact is in in-play betting, also known as live betting. Users are able to place bets during a game based on real-time updates. Companies like DraftKings are investing heavily in this space, even acquiring other businesses to enhance their ability to offer these fast, data-driven betting options.
We asked Artificial Intelligence to predict the AFC Championship game… here’s what they said: pic.twitter.com/OPaIjbazZz
— DraftKings (@DraftKings) January 28, 2023
SharpLink GamingBy analyzing your betting history and preferences, AI is now able to suggest wagers and notify you about games that align with your interests. SharpLink Gaming, for example, uses the tech to tailor the sportsbook experience, learning from how users interact with the platform to offer more relevant recommendations.
SportradarBeyond improving the betting experience, it also plays a big role in fraud detection and risk management. Because these systems can process huge volumes of data quickly, they’re well-suited to spotting unusual patterns that might suggest suspicious activity or match-fixing. Sportradar is one company that uses AI to monitor for these kinds of issues, helping to maintain the integrity of sports betting.
And then there’s generative AI, which helps improve user engagement by generating personalized content and suggestions. For instance, Stats Perform uses generative AI to make betting recommendations that feel more interactive and tailored to the user.
‘False sense of precision’However, industry experts have pointed out that the rise of AI in this space comes with both opportunities and serious caveats.
Speaking to ReadWrite, Tomi Huttunen, senior manager at Online Casino Suomi (OCS), confirms that AI has become a widely adopted tool among bettors. “People are more tech-savvy than ever,” he explains. “Tools like ChatGPT, Gemini, and custom-built prediction models are now being used to analyse team form, historical data, player performance, and even social media sentiment.” He notes that while gambling has always relied on data and statistics, AI gives users the ability to process more information, faster, than ever before.
That speed and perceived edge over traditional bookmakers are a big part of AI’s appeal. “AI offers speed and scale,” Huttunen adds. “A single user can process more data in 10 seconds with AI than a team of analysts could in hours. There’s also an increasing belief that AI can ‘outsmart the bookies.’”
But he’s quick to caution users against blind trust. “Models like ChatGPT aren’t designed to give betting advice,” he says. “They’re designed to generate language based on patterns. So while they can summarise stats and trends well, they don’t have access to live odds, injury updates, or market movement in real time. That creates a false sense of precision for users who treat the outputs like gospel.”
He also raises concerns about the lack of regulation around AI-generated tips, which could lead to risky betting behaviors if not handled responsibly. “Transparency and education will be key,” he warns.
‘AI as a tool, not an oracle’Robert Kraft, CEO and co-founder of Atlas World Sports, echoes the sentiment that AI has gained serious momentum in the betting world.
According to Kraft, the rise in interest is fueled by three main factors: the massive availability of data, the speed at which AI can process complex variables, and the growing accessibility of AI tools to everyday users. He states that “even casual bettors [can] leverage advanced analytics.”
Still, he stresses that AI predictions are not foolproof. “The unpredictability of sports (e.g., referee calls, locker-room dynamics) still limits AI/ML’s accuracy,” he tells us. Kraft is particularly concerned about overreliance on general AI tools like ChatGPT and Gemini. “ChatGPT and Gemini aren’t designed for sports betting. They may generate plausible-sounding but statistically unvetted predictions, leading users astray.”
“AI could exacerbate problem gambling if marketed as ‘infallible.’ Responsible gambling requires transparency about AI’s limitations.” – Robert Kraft, Atlas World Sports CEO
Bias in training data and the fast-changing nature of sports are other challenges Kraft showcases. “Let’s also keep in mind, players age, teams change coaches, General Managers, etc. so Machine Learning cannot simply take last season’s model and extrapolate 100% of that model with just a few new inputs, so it really must be robust and built with the understanding that each season is almost a different data set completely while using the historical models as a base-line of sorts,” he explains.
For Kraft, the best results come from hybrid models that combine AI’s data processing power with human judgment. “We come from an agnostic data perspective and see AI as a tool, not an oracle,” he says.
Adam Small, cofounder of CasinoReports.com, also sees a clear rise in AI-generated betting content. “While I don’t know that there are numbers publicly available on this, in general, there is a huge rush of AI-generated content in all sectors, particularly where there’s money to be made,” he says.
According to Small, many sites are now producing betting articles with little or no human input. Tools like ChatGPT make it easy to generate persuasive arguments for betting picks, even in the voice of a famous sports personality.
He’s not particularly concerned about AI’s ability to predict outcomes. “Predictions are a dime a dozen,” Small says. “There is a tiny fraction of the aforementioned ‘everyone’ that is actually profitable to listen to. I’m skeptical Gemini or ChatGPT would be any worse at picking outcomes than the typical analyst people read for betting advice.” His biggest worry is around misuse, such as AI content impersonating real people, but he believes most issues are manageable with proper disclosure.
The issues with AI in sports betting predictionsBesides the fact that bots can sometimes make up information, there’s also a big risk when it comes to sharing financial details with AI tools. Any time you type in things like your betting balance, payment method, or favorite platforms, you could be exposing sensitive information.
In 2023, ChatGPT actually had a data breach that revealed personal details like social security numbers, email addresses, and locations.
In a statement provided to ReadWrite, Christoph C. Cemper, founder of AI prompt management company AIPRM, said: “OpenAI’s terms and conditions state they may retain your data, including information you’ve chosen to share during your chats, in order to help train their future models.
“Their terms explain: ‘We may use the data you provide us to improve our models.’ It’s advised to be selective about what you type. If you don’t want it repeated to other users in the future, you should avoid sharing it with ChatGPT.”
bet365’s terms clearly outline prohibited activities, including arbitrage strategies, VPN use, and unauthorized third-party account access. Credit: bet365
A lot of people also don’t realize that many of the biggest betting sites actually ban the use of third-party software, including AI tools. For example, bet365’s terms and conditions clearly state that they can void your bets, withhold any winnings, and even suspend or close your account if you’re caught using prohibited software.
On top of that, you could be held responsible for any losses the company faces because of your actions.
That said, AI’s ability to process huge amounts of data quickly and make that information accessible to just about anyone it’s changing how people look at games and place their bets. But even though the technology is impressive, it’s far from perfect and not a magic solution for winning every time.
Featured image: Canva
The post Why the rise in AI sports betting predictions comes with serious risks appeared first on ReadWrite.