Based on testing, the companies behind Ike Tech claim this process has a 100 percent success rate in age verification, more or less calling the tech infallible. “The FDA told us it’s the holy grail technology they were looking for,” Wang says. “That’s word-for-word what they said when we met with them.” The FDA did not respond to WIRED’s request for comment.

But Glantz is not at all convinced these protections will work.

“The FDA is just showing their pro-industry bias,” Glantz says. “If I were running the FDA, I would prohibit these devices from having any Bluetooth capability at all, period. There are just too many ways it could go south. Every technical fix has a work-around.”

The verification features would be tied to just one person, so when the vape is on, that person could share a puff with anyone nearby without verifying their age. At that point, Wang says it comes down to personal responsibility.

“You really have to count on the responsibility of that person,” Wang says. “If it’s a 21-year-old or older person, of course, that’s fine, but if you really want to hand it to an underage person, then you are really irresponsible.”

Wang says the goal is to implement additional features in the verification process, like geo-fencing, which would force the vape to shut off while near a school or on an airplane. In the future, the plan is to license this biometric verification tech to other e-cig companies. The tech may also grow to include fingerprint readers and expand to other product categories; Wang suggests guns, which have a long history of age-verification features not quite working.

Vapor Ware

The time frame for when Ike Tech might actually be out in the world—and how much it will cost when added to vape cartridges—is still hazy. Wang says there are already partnerships with two nicotine companies, but won’t say which or when that will emerge. “In 2026, there will be a clear indication of when our solution will be approved and how many other brands will license our technology.”

Wang’s ideal version of a vape, he posits, would be a safe, clean way to inhale nicotine.

“In the industry, we have a saying: ‘Nicotine never killed a single person,’” Wang says. “To a large degree, e-cigarettes are a safer way to consume nicotine.”

Glantz rejects that notion by bringing up practices like “smoking topography,” where nicotine companies track how smokers puff the product differently, then control how much nicotine is dispersed at a time to maximize the addictive potential.

He also takes issue with the fundamental problem that e-cigarettes and vapes are cleaner than traditional cigarettes at all. While the problem with cigarettes and cheap vapes might be the other chemicals, nicotine itself is not a harmless substance.

“You can’t make a healthy e-cigarette; it’s impossible,” Glantz says. “It’s true that nicotine isn’t a carcinogen, but it has all kinds of adverse cardiovascular effects. Nicotine screws up your nervous system.”

For nicotine to be absorbed as vapor, it has to be reduced to ultrafine particles. That’s what the heating does, and those particles can have all sorts of adverse health effects.

“There are all of these other implications that are extremely serious that nobody’s really thinking about,” Glantz says. “Even if the age-verification thing worked, it’s still not worth it.”

Streaming was supposed to simplify everything. Instead, it is starting to look a lot like the cable bundles it replaced, with higher monthly costs, more fragmentation, and ads creeping back into tiers that once promised otherwise. For a growing number of viewers, that frustration is pushing them back toward something far less convenient but far more reliable: physical media.

Owning your movies and TV shows means they do not disappear when licensing deals expire or platforms reshuffle their catalogs overnight. You press play, and it works. No buffering, no logins, no surprise paywalls. It also means consistently better performance. DVD, Blu-ray, and especially 4K Ultra HD Blu-ray still deliver higher bitrates than streaming, which translates into cleaner images, more stable motion, deeper color, and lossless audio formats that actually take advantage of a proper surround sound system.

Finding discs is not as easy as it used to be. Major retailers like Best Buy and Target have scaled back or eliminated in store selections, but physical media is not disappearing. It is shifting. Walmart and Barnes & Noble still carry a solid range in store, and online options are stronger than ever. Boutique labels like The Criterion Collection, Kino Lorber, and Arrow Video continue to release restored editions, special features, and curated catalogs that streaming platforms rarely match.

This article and the three companion videos break down exactly what you get from DVD, Blu-ray, and 4K Ultra HD Blu-ray. Not just resolution numbers, but the real world differences in picture quality, HDR support, audio formats, and overall viewing experience so you can decide if going back to discs is nostalgia or just common sense in 2026.

The Bottom Line

DVD still has a place, but Blu-ray and 4K Ultra HD Blu-ray remain the standard for anyone who actually cares about how movies look and sound. Streaming has improved, but it still cannot match the higher bitrates, consistent image quality, and lossless audio you get from a disc spinning in a dedicated player. That gap becomes obvious the moment you pair them with a modern 4K TV and a real surround sound system.

The three videos break down what matters in 2026. DVD is limited to standard definition. Blu-ray delivers full HD with excellent clarity and solid audio. 4K Ultra HD Blu-ray adds true 4K resolution, HDR formats like HDR10 and Dolby Vision, and the best available home audio formats. To take advantage of it, you need the right hardware. A 4K Ultra HD Blu-ray player, a compatible TV with HDR support, and an audio system that can handle formats like Dolby Atmos or DTS:X. Get that right, and physical media is not just surviving. It is still the benchmark.

Where to buy?

Related Reading:

Samsung is widening the reach of its next big Android update, with One UI 8.5 beta now rolling out to a wave of older Galaxy devices. This is not just rolling out to its latest flagships.

After spending months in testing on the Galaxy S25 series, the beta programme is expanding to include last year’s premium phones, foldables and even tablets. Newly supported devices include the Galaxy S24 lineup, Galaxy Z Fold 6 and Flip 6.

In addition, the Galaxy Tab S11 range is supported as well. This is a clear sign Samsung is speeding up its software rollout strategy.

As AI and robotics become an ever-more important presence in everyday life, the use cases are quickly going from science fiction to real life.

One of the most popular areas of interest is autonomous vehicles, self-driving cars able to navigate the roads and get us to our destination without needing to touch the wheel.

Modern fraud attacks look like a relay race where different tools and actors handle each stage of the journey from signup to cash-out.

When you only inspect one signal at a time, such as IP or email, attackers simply shift to a different part of the chain and still succeed.

Anatomy of a Modern Fraud Chain

A typical attack chain starts with automation to create scale. Attackers use bots and scripts to open large numbers of accounts with minimal human effort, often rotating infrastructure to avoid rate limits and simple bot rules.

Those bots are usually powered by “aged” or compromised emails and leaked credentials so that every account looks like it belongs to a long standing user instead of something created yesterday.

Residential proxies then mask traffic behind real consumer IP ranges, making traffic appear like normal home users rather than data centers or known VPN services.

Once those accounts are established, they shift tactics from automation to slower, human driven sessions to blend into normal usage.

At this point the chain reaches account takeover and monetization, using malware links, phishing, and credential stuffing outputs to log in, change details, and push through high value transactions.

Throughout this lifecycle, the tools are mixed and matched. A single actor may move from a headless browser and proxy at signup to a mobile device emulator and different proxy provider at login, then hand off access to another party who specializes in draining funds or exploiting promo campaigns.

This is exactly why a point in time, single signal check rarely tells the full story

False Positives from Siloed Checks

When teams lean on one dominant signal, such as IP reputation, false positives become a daily problem. Legitimate users on shared Wi Fi, mobile carrier NATs, or corporate VPNs can inherit the poor reputation of a small number of bad actors on the same ranges, even though their intent is clean.

Blocking by email alone has similar issues, since free webmail domains are used by both sophisticated attackers and completely normal customers.

Identity centric controls on their own also hit a wall. Static data checks, like simple name and document matches, are easy to spoof for synthetic identities built from real data fragments.

Device centric controls that only look for rooted phones or emulators can miss fraudsters operating on seemingly normal devices that have been compromised earlier in the chain. Even bot specific solutions can create blind spots when they work alone.

Once a credential stuffing campaign ends and attackers pivot to manual logins with the same stolen credentials, pure bot tools see only “human” traffic and approve it. The result is a pattern where high risk users are blocked while determined adversaries adapt and slip through.

Multi-Signal Correlation in Practice

Effective fraud defense comes from correlating IP, identity, device, and behavioral signals at every step of the journey instead of evaluating each one in isolation.

An IP that looks slightly suspicious on its own becomes clearly abusive when tied to dozens of new accounts on the same device fingerprint and similar behavioral patterns during the first session.

Likewise, a user with an apparently normal device and clean email reputation can still be high-risk if login behavior reflects credential stuffing patterns or access follows known malware distribution campaigns.

Modern decision engines improve accuracy by weighing hundreds or thousands of data points together rather than enforcing rigid rules on a single attribute.

For organizations, that means unifying what were once separate views. IP intelligence, device fingerprinting, identity verification, and behavioral analytics should feed the same risk model so that each event is scored in context, not as a disconnected log line.

This multi signal approach is the most reliable way to raise the bar for attackers while reducing friction for genuine customers.

Case Study: Stopping Coordinated Signup Abuse

Consider a self service SaaS platform that offers a generous free tier and trials. As the product grows, abuse appears in the form of thousands of signups used to scrape data, test stolen cards, or resell access under the radar.

Early countermeasures rely on blocking certain IP ranges and obvious disposable email domains, but this only dents the problem and begins to impact small teams and freelancers on shared networks.

By shifting to a multi-signal model, the platform starts scoring signups across IP, device, identity, and behavior together.

New accounts that reuse the same device fingerprint with different emails, come from IPs recently seen in automated traffic, or immediately exhibit scripted behavior are grouped into coordinated abuse clusters instead of being evaluated one by one.

This lets the team apply precise responses, such as challenging only high risk clusters with additional verification or silently limiting their capabilities while letting low risk signups proceed without friction.

Over time, feedback from confirmed abuse and confirmed good users trains the scoring model, driving down false positives while pushing organized attackers to spend more effort for less return.

Outpacing Fraud Trends

Attackers are no longer tied to a single tool or weak point in your stack. They combine proxies, bots, synthetic identities, leaked credentials, and malware infrastructure across multiple stages, which means that single signal defenses will always lag behind.

To keep pace, fraud teams need correlation across IP, identity, device, and behavior in one coherent risk view rather than a collection of disconnected checks.

From here, the conversation shifts to how to operationalize that unified model, integrate it into existing workflows, and measure its impact on both loss reduction and customer experience.

Book a free consultation with a fraud expert today!

About IPQS

IPQS is a founder-led, self-funded company built on a simple principle: fraud prevention should be driven by real intelligence and a multi-layered approach. From day one, we’ve focused on owning the full lifecycle of our technology—developing and maintaining our own global data network, honeypots, and fraud intelligence specialists. This approach gives our customers a distinct advantage: faster insights, greater accuracy, and complete transparency into how decisions are made. By staying independent, we prioritize long-term innovation over short-term gains, continuously evolving our platform to stop fraud before it starts.

Sponsored and written by IPQS.

The next era of software ‘will be built for agents, by agents’, said Jentic CEO Sean Blanchfield.

Irish AI start-up Jentic is jumping on the OpenClaw frenzy with ‘Jentic Mini’, a free, open source, self-hosted offering for developers building with the product.

The months-old OpenClaw project has taken the developer world by storm, but security issues associated with letting personal agents access computers has already led to a number of different iterations from big-name brands such as Nvidia, with its open source stack NemoClaw, and Anthropic, which recently integrated OpenClaw’s text features into Claude.

Jentic Mini also markets itself to be safer to use in real-world software environments. Jentic said that the API execution layer gives developers a better way to govern how agents access tools, APIs and workflows to help reduce risks that come with broad and unmanaged credential exposure.

Jentic Mini is available on the start-up’s website, as well as on GitHub. It can also be used with other general-purpose agents including NemoClaw.

With more than 10,000 APIs, agents created using Jentic Mini will come with a “machine-usable map” of the tools and workflows that can be used, while enabling a more structured and controlled way to connect agents to real systems, the company said. Jentic’s standard model is available as a verified connector in Claude.

“The next era of software will not be built for humans. It will be built for agents, by agents,” said Sean Blanchfield, the CEO and co-founder of Jentic.

“Jentic Mini gives developers a free, open source foundation for that shift, connecting general-purpose agents to real systems through an AI-curated catalogue of more than 10,000 APIs and workflows. We want to make it dramatically easier to deploy agents that do real work.”

Jentic became the first Irish company to be selected for the AWS generative AI accelerator last year. The selection followed a €4m pre-seed raise in 2024, one of the largest in Ireland that year.

Don’t miss out on the knowledge you need to succeed. Sign up for the Daily Brief, Silicon Republic’s digest of need-to-know sci-tech news.