Fraud protection is a race against scale. 

For instance, Mastercard’s network processes roughly 160 billion transactions a year, and experiences surges of 70,000 transactions a second during peak periods (like the December holiday rush). Finding the fraudulent purchases among those — without chasing false alarms — is an incredible task, which is why fraudsters have been able to game the system. 

But now, sophisticated AI models can probe down to individual transactions, pinpointing the ones that seem suspicious — in milliseconds’ time. This is the heart of Mastercard’s flagship fraud platform, Decision Intelligence Pro (DI Pro). 

“DI Pro is specifically looking at each transaction and the risk associated with it,” Johan Gerber, Mastercard’s EVP of security solutions, said in a recent VB Beyond the Pilot podcast. “The fundamental problem we’re trying to solve here is assessing in real time.”

How DI Pro works

Mastercard’s DI Pro was built for latency and speed. From the moment a consumer taps a card or clicks “buy,” that transaction flows through Mastercard’s orchestration layer, back onto the network, and then on to the issuing bank. Typically, this occurs in less than 300 milliseconds. 

Ultimately, the bank makes the approve-or-decline decision, but the quality of that decision depends on Mastercard’s ability to deliver a precise, contextualized risk score based on whether the transaction could be fraudulent. Complicating this whole process is the fact that they’re not looking for anomalies, per se; they’re looking for transactions that, by design, are similar to consumer behavior. 

At the core of DI Pro is a recurrent neural network (RNN) that Mastercard refers to as an “inverse recommender” architecture. This treats fraud detection as a recommendation problem; the RNN performs a pattern completion exercise to identify how merchants relate to one another. 

As Gerber explained: “Here’s where they’ve been before, here’s where they are right now. Does this make sense for them? Would we have recommended this merchant to them?” 

Chris Merz, SVP of data science at MasterCard, explained that the fraud problem can be broken down into two sub components: A user’s pattern behavior and a fraudster’s pattern behavior. “And we’re trying to tease those two things out,” he said. 

Another “neat technique,” he said, is how Mastercard approaches data sovereignty, or when data is subject to the laws and governance structures in the region where it is collected, processed, or stored. To keep data “on soil,” the company’s fraud team relies on aggregated, “completely anonymized” data that is not sensitive to any privacy concerns and thus can be shared with models globally. 

“So you still can have the global patterns influencing every local decision,” said Gerber. “We take a year’s worth of knowledge and squeeze it into a single transaction in 50 milliseconds to say yes or no, this is good or this is bad.”

Scamming the scammers

While AI is helping financial companies like Mastercard, it’s helping fraudsters, too; now, they’re able to rapidly develop new techniques and identify new avenues to exploit.  

Mastercard is fighting back by engaging cyber criminals on their turf. One way they’re doing so is by using “honeypots,” or artificial environments meant to essentially “trap” cyber criminals. When threat actors think they’ve got a legitimate mark, AI agents engage with them in the hopes of accessing mule accounts used to funnel money. That becomes “extremely powerful,” Gerber said, because defenders can apply graph techniques to determine how and where mule accounts are connected to legitimate accounts. 

Because in the end, to get their payout, scammers need a legitimate account somewhere, linked to mule accounts, even if it’s cloaked 10 layers down. When defenders can identify these, they can map global fraud networks.

“It’s a wonderful thing when we take the fight to them, because they cause us enough pain as it is,” Gerber said. 

Listen to the podcast to learn more about: 

• How Mastercard created a “malware sandbox” with Recorded Future; 

• Why a data science engineering requirements document (DSERD) was essential to align four separate engineering teams;

• The importance of “relentless prioritization” and tough decision-making to move beyond “a thousand flowers blooming” to projects that actually have a strong business impact;

• Why successful AI deployment should incorporate three phases: ideation, activation, and implementation — but many enterprises skip the second step. 

Listen and subscribe to Beyond the Pilot on Spotify, Apple or wherever you get your podcasts.

If you’re used to hopping between gaming servers, sharing memes, or catching up with friends on Discord, be ready for a small twist in how the platform works. Starting March 2026, Discord will introduce a global age verification system that automatically places every user into a “teen-appropriate experience” until they prove they’re an adult. To fully unlock age-restricted content and certain features, users will need to verify their age via face-based AI estimation or a government-issued ID.

According to Discord’s official announcement, this change is part of a broader safety push to control access to adult or graphic material and comply with international safety expectations. But before you panic that Discord is turning into a DMV, it’s worth noting what changes and what stays the same, if you don’t verify right away.

What works and what doesn’t without age verification

Under Discord’s upcoming teen-by-default setup, everyday features won’t suddenly vanish. Users can still send messages, chat in most regular servers and DMs, hop into voice calls with friends, and hang out in their usual gaming communities just like before.

However, anything marked as adult or age-restricted will stay locked behind a verification wall. That includes access to certain servers and channels, speaking privileges in Stage channels, and viewing unfiltered or sensitive content, which will remain blurred by default. Some message request controls and notification customizations may also stay limited until age verification is completed, meaning full access only unlocks once you confirm you’re old enough.

Discord knows this move isn’t going to earn universal applause. Any time an app asks for a face scan or ID, a few users are bound to hit the exit button instead. Still, the company says the bigger picture is safety first, aiming to set a more secure, teen-friendly default for everyone rather than letting the wild west stay wild.

The good news is your everyday Discord life isn’t getting kicked from the party. DMs, regular servers, and voice chats with the squad will keep humming along. But if you want the full, no-filters version of Discord, complete with adult servers, spicy memes, and every corner of your late-night gaming hangouts, you’ll have to show you’re old enough to enter. Think of it less like a ban and more like a digital bouncer asking for ID at the door.

Sony has all but confirmed what the leaks have been hinting at for months: a new pair of flagship earbuds is on the way.

The company has released a short teaser video for the WF-1000XM6, confirming that its next-generation wireless buds are “coming soon”. In addition, a full reveal is set for February 12 at 8am PT.

While the teaser doesn’t give much away on paper, it does make it clear that Sony is getting ready to refresh one of the most respected noise-cancelling earbud lines around. The WF-1000X series has long sat at the top of Sony’s audio lineup. Now, the XM6 looks set to continue that run.

The video itself only briefly shows the earbuds in shadow, but by this point the design isn’t exactly a mystery. Full renders have already leaked, suggesting a refined look rather than a dramatic redesign. Expect a similar premium feel, with subtle tweaks aimed at comfort and wearability rather than a radical change in direction.

As ever, active noise cancellation is expected to be the headline feature. Sony’s ANC performance has consistently ranked among the best in the business, and there’s little reason to think the WF-1000XM6 will be any different. While Sony hasn’t confirmed specific specs yet, improved processing and smarter noise handling feel like safe bets for a generational update.

YouTube TV announced plans to introduce new genre-specific bundles last year to give users the flexibility to pick and pay for the content they actually want to watch. Four of these bundles are now live in the US, with the platform planning to roll out over ten bundles across Sports, News, Entertainment, and Family content in the coming weeks.

YouTube TV announced the rollout in a recent blog post, highlighting the content included in the introductory bundles and their pricing. The base Sports plan, priced at $64.99/month, will give subscribers access to all major broadcasters, alongside sports networks like FS1, NBC Sports Network, and all of ESPN networks. Subscribers ill also gain access to ESPN Unlimited this fall.

The Sports + News plan includes everything from the Sports plan, along with national news networks such as CNBC, Fox News, MSNBC, CNN, CSPAN, Bloomberg, and Fox Business. It’s priced at $71.99/month. The Entertainment plan, which is the most affordable of the four, costs $54.99/month and includes all major broadcasters and content, “ranging from FX dramas to Hallmark classics, with channels such as Comedy Central, Bravo, Paramount, Food Network, HGTV, and many more.”

Finally, the News + Entertainment + Family plan, priced at $69.99/month, combines news and entertainment channels with family content like Disney Channel, Nickelodeon, National Geographic, Cartoon Network, PBS Kids and more.

YouTube TV bundles offer big savings, especially for new subscribers

All four bundles is priced lower than the main YouTube TV plan, the platform’s most comprehensive offering with over 100 networks across genres. For those who haven’t previously subscribed to YouTube TV, the bundles are available at a discount.

The Sports plan is available for $54.99 per month for the first year, while the Sports + News, Entertainment, and News + Entertainment + Family plans are priced at $56.99, $44.99, and $59.99/month, respectively, for the first three months.

• EWS doesn’t deliver security, scale and reliability to today’s standards
• It will be shut down for cloud environments from April 2027
• ‘Scream tests’ will help show any dependencies

Microsoft has confirmed it will be phasing out Exchange Web Services (EWS) for Microsoft 365 and Exchange Online after nearly two decades of services, and we’ve been given all the key dates.

As soon as October 2026, the company will disable EWS by default for Exchange Online tenants, with the final shutdown set for April 1, 2027.