When you think of Memorial Day sales, you probably think of mattresses and other home goods. And while those items are definitely discounted, now is also a good time to purchase tech. Personally, I’m not buying anything right now unless it’s discounted—and fortunately many of our top picks are. Whether you’re shopping for a power bank, a new pair of headphones, or some other gadget, I’ve rounded up the best Memorial Day deals for your perusal. Most of these deals end at the end of the day.

Check out our buying guides for more recommendations, including the best headphones, the best laptops, and the best cheap phones. You might also want to check out our additional Memorial Day deals coverage.

Updated Monday, May 25: We’ve checked prices, removed expired deals, added 6 new deals, and ensured accuracy throughout.

WIRED Featured Deals:

Sony WH-1000XM5 for $248 ($152 off)

The Sony WH-1000XM5 have a very frustrating name, but they’re the predecessor to our favorite wireless headphones, and they’re still an excellent pick if you don’t want to shell out for the new WH-1000XM6. They go on sale frequently, but rarely drop this low in price, which comes within $5 of their all-time low. If you’re in the market for over-ear headphones, they’re hard to beat. They’re comfortable, portable, lightweight, and stylish, and they’ll make your music sound great no matter what you like to listen to.

• AI-powered hackers now exploit software flaws faster than companies can patch systems
• Mobile phishing scams now outperform traditional email attacks across corporate environments worldwide
• Unauthorized AI tools are quietly leaking sensitive company information across global workplaces

For the first time in nearly two decades, exploiting software vulnerabilities has overtaken stolen passwords as the primary way hackers breach corporate networks.

Verizon’s 2026 Data Breach Investigations Report claims the exploitation of vulnerabilities now accounts for 31% of all confirmed data breaches.

Over the past two decades, technical debt meant outdated architecture, messy code, and poorly maintained documentation. That definition is no longer sufficient in the AI era, where failure modes are more subtle and often non-linear. AI systems are introducing new layers of technical debt that live across prompts, models, and data dependencies — making these layers less visible, harder to measure, and often more dangerous than traditional debt.

A crisis hiding in plain sight

The complexities of AI systems and their associated failures have been well documented. A 2025 MIT study found that 95% of AI projects fail to reach production or deliver value. A similar study by S&P Global Market Intelligence found that 42% of businesses scrapped multiple AI initiatives in 2025 — a sharp increase from 17% the previous year. Various reasons are cited for these failures, but most of them point to poorly designed and implemented systems that are complex to manage and have multiple hard-to-monitor failure points, leading to a rapid accumulation of AI debt. 

Traditional technical debt was localized to the codebase, and bugs were usually easily reproducible. Consequently, bugs could be easily identified during tests and fixed through rearchitecting the codebase. However, AI debt is much more distributed, manifesting across prompts, models, data pipelines, and all associated infrastructure. It is also more intermittent: Due to the probabilistic nature of AI, systems do not always respond the same way, leading to intermittent failures. This makes it much more challenging to identify risks during testing, and also creates a need for more continuous monitoring even post-deployment to prevent gradual drift and worsening performance.

The new forms of AI debt

AI debt typically manifests across four new forms, each of which comes with its own set of risks.

Prompt debt is the most visible of these. A modern version of ‘spaghetti code,’ this can include undocumented prompt tweaks, accumulated ‘quick-fix’ prompts that lead to inconsistencies, neglected version control of prompts, and ‘prompt stuffing’ (the cramming of extraneous data or context directly into AI prompts). All these combine to make prompts a form of untyped, untested code without any version control, leading to increased brittleness and vulnerabilities.

Model dependency debt is another increasingly common form of AI debt. Most enterprises now depend on a mixture of external models developed by leading foundation model providers; applications and agents are built on top of API calls to these models. Consequently, application logic now depends on models that are external to the core system, and that cannot be clearly controlled. As models update, performance varies and reproducibility is lost — prompts tuned for one model may fail or perform poorly when switched to another model, whether an update from the same provider or from another provider.

Most enterprise AI deployments today use retrieval-augmented generation (RAG), which pulls in additional context from enterprise data repositories. Retrieval debt is a consequence of these repositories having messy data, duplicated documents, and outdated information. This causes AI to return technically correct answers that are outdated and no longer relevant, causing downstream failures. Unlike hallucinations, these are harder to detect because they were correct, perhaps even until recently, and hence look correct to any tester. 

Evaluation debt reflects the lack of standardization in testing and monitoring for AI models and applications. While AI benchmarks exist, they tend to focus on narrow tests and reflect point-in-time results. Most enterprises lack consistent testing standards, ground truth datasets, and real-time monitoring of deployments; there is no equivalent yet of continuous integration /continuous delivery (CI/CD) for prompts. As a consequence, CIOs and CTOs do not have clear visibility into model performance and cannot track improvements or worsening of models. 

All of these are in addition to traditional forms of technical debt, which still manifest across the tools and systems that AI applications and agents interact with, read from, or write to. A rapid increase in the adoption of AI-generated code (often deployed without inadequate testing) is further aggravating inconsistencies within, and poor maintainability of traditional codebases. 

The new forms of AI debt combine with these earlier forms of technical debt to compound rapidly and create large-scale risks that can cause catastrophic failure of entire enterprise deployments. Solving for these risks is made even more challenging by the distributed nature of AI ownership – most systems span engineering, product, data, and business teams, leading to unclear accountability when an error is identified. 

As a result, these risks manifest in the form of escalating compute costs, inaccuracies in AI outputs, and increasing exceptions that need to be handled by humans — leading to projects often stalling and failing due to unclear return-on-investment stories and a lack of trust from users. 

How enterprises can prevent AI debt

AI debt will not be solved by ‘better’ models — failure rates remain high despite models already having high accuracy. The solution to AI debt requires better system design, integration, controls, and changes in organizational culture. 

First, prompts need to be treated as code. This involves careful version control, documentation, and rigorous testing both pre- and post-deployment for all possible prompt configurations. Best practices from the traditional world of coding — such as the use of smaller prompt blocks instead of large prompt-stuffed walls, or reducing the use of hard-coded parameters — can also help mitigate AI debt. 

Second, evaluation needs to be built into the entire AI infrastructure stack. Continuous evaluation pipelines need to be established and must reflect a wide variety of metrics measuring both technical and business-aligned metrics. In addition, AI observability systems should be integrated to monitor output quality, failure rates, model drift, and data drift.

Third, explainability should be included by default in all AI results to make up for limited reproducibility. Data lineage, models used, and the steps followed should be clearly traceable so as to allow auditability of results and correction in case of any systemic errors. 

This requires explicit AI debt reduction programs and associated budgets, similar to earlier waves of investment in security or in cloud modernization. These need to be driven at a CXO level by key leaders to prevent costly rework later.

Conclusion: A stitch in time

Enterprise AI deployments are not just static code; they are living systems that interact with the entire enterprise stack. As a result, the defining challenge in an agentic enterprise will not be building or deploying intelligent systems, it will be maintaining these systems to ensure continued reliability during real-world operation.

Enterprises that seek to proactively identify and mitigate AI debt from the design phase itself are the likeliest to build sustainable AI platforms that deliver significant long-term productivity boosts across the organization. 

Vikram is a principal at Cota Capital, where he invests in early-stage enterprise tech and deep tech companies.

Welcome to the VentureBeat community!

Our guest posting program is where technical experts share insights and provide neutral, non-vested deep dives on AI, data infrastructure, cybersecurity and other cutting-edge technologies shaping the future of enterprise.

Read more from our guest post program — and check out our guidelines if you’re interested in contributing an article of your own!

• Kansas City schools replace 30,000 Windows and Chromebooks with Apple devices
• Concerns raised over financial loss from retiring functional school computers
• District cites security, durability, and “student pride” as reasons for the Apple switch

The Kansas City Public Schools district has announced a sweeping transition which will remove tens of thousands of non-Apple devices from its classrooms.

According to information on the district’s website, administrators will replace more than 30,000 Windows PCs and Chromebooks with Apple hardware over the coming months.

Laptops have always come in all shapes and sizes, but recently I’ve seen ultrabooks become super popular. Maybe it’s just me who’s been drowning in work with my other colleagues, but wherever I go, I only see these thin and light laptops. A big part of this push can be attributed to efficiency gains, which mean we can fit more power into these chassis and get serious performance out of them. There’s one brand that’s always bet big on ultrabooks, and that’s Asus. Their ZenBook lineup is probably what comes to mind when anyone thinks about a serious laptop for professionals, at least in India.

I’ve tested several of these ZenBooks in the past, and they’ve always passed my tests with flying colors. However, 2026 has been proving to be a big challenge with rising memory prices driving costs up and value down. The latest iteration of the ZenBook 14 landed on my table a couple of weeks back. Asus has refreshed it with the latest Ryzen processor and a new OLED touch display option. To do the laptop justice, I switched my MacBook (it’s easy, my work is on Chrome, mostly) and made the ZenBook 14 UM3406GA my daily driver. Here’s how it stacks up in 2026.

Design & Hardware

If you remember the last ZenBook 14 (which we reviewed previously), then the new iteration would feel right at home. The laptop is available only in stealth black, and I’m a fan. It’s super elegant, and I’m a bit biased towards stealthy machines. The ZenBook 14 would fit perfectly for a corporate meeting or a quick stop at the nearby cafe to catch up on some work. Sadly, if you like experimenting with colors, there’s no other option. For some fun, the Vivobook lineup should come in handy.

The new Asus logo adds to the premium appeal, and everyone whom I gave the laptop to loved the design. There’s just one small problem, though: fingerprints. Asus has yet to fix this problem, as the black finish picks up sweat and grease from your palms and leaves them printed on the surface unless you use a cleaning cloth. It’s not a major problem, but something worth mentioning.

Minor complaints aside, what the Asus ZenBook 14 does best is portability. The 14-inch form factor is best for people who are constantly on the go but still need enough screen space to multitask. To put this to the test, I took the laptop on a recent work trip to Delhi, where the 1.2 kg weight didn’t put much stress on my shoulders. I took it out to work at the airport, and everything went fine without hiccups. The metal construction meant I wasn’t worried about putting too much pressure on the backpack. I also tested the build myself and observed no flex in either the keyboard deck or the display panel. The hinge holds the display firmly enough, though it’s a little stiffer than I’d like. It lifts the laptop ever so slightly, which can be annoying, especially coming from a MacBook.

As far as ports are concerned, Asus has your back. The left side houses a couple of Type-C ports, with one supporting USB 4.0 Gen 3 (40 Gbps data bandwidth, DisplayPort, and Power Delivery). Beyond that, there’s a full-size HDMI 2.1 port, an audio combo jack, and a USB 3.2 Gen 1 Type-A port on the left. I had no problems connecting accessories to the laptop.

Keyboard & Trackpad

A good keyboard is one of the cornerstones of an ultrabook, since most people spend hours typing documents (like me writing this review). Fortunately, the one on the ZenBook 14 is quite good. Coming from a MacBook, my keyboard standards aren’t that high, and I got used to this one quickly. The keyboard is generously spaced, so there’s no cramped action. Typing feels clicky enough, and there’s good feedback in the end. Backlit support is present, and, thankfully, Asus hasn’t gone with gray keycaps for contrast, which would make them visible at night.

The same praises can be carried to the touchpad as well. Sure, it’s not the haptic one I’m used to, but I’ve seen plenty of people who love physical touchpads. If that’s you, you’re in luck: the one on the ZenBook 14 is quite large, and I didn’t find any dead zones. Instead, Asus has bundled a few extras with the trackpad, including a light-up number pad that turns on when you press the button on the top right.

Display & Camera

Last year’s Asus ZenBook 14 came with arguably the best display ever, with a 3K OLED 120Hz panel. This year, though, Asus has decided to switch things up. The new one gets a 14-inch FHD+ OLED panel with a 60Hz refresh rate. If that sounds underwhelming, there is now support for touch. This adds a whole new dimension to using the laptop, and I’d much rather have this functionality over the tad bit of extra sharpness. Beyond that, watching content is an absolute breeze on the ZenBook 14. I was catching up on Better Call Saul (I know I’m late), and the 100% coverage of the DCI-P3 color space kept everything stunning with vibrant yet natural colors. The contrast was excellent, and I could make out the different details on the faces.

The ZenBook 14 is also VESA DisplayHDR True Black 500-certified, so HDR content should pop out a bit. In everyday use, I measured a peak of around 450 nits, which is plenty for working in a bright cafe or on a cloudy day outdoors.

As for the camera, the 1080p webcam performs as you might expect. The sensor is sharp enough not to make the videos look muddy, but it struggles in low-light environments, with noise creeping in. Aside from the basics, Asus has bundled several AI features, including 3D noise reduction and AI Noise Cancellation. Both of these work fine, and I actually enjoyed my time giving presentations on the laptop.

Performance & Gaming

Performance is another pillar of the ultrabook experience since nobody wants to deal with an underpowered processor that hangs up during an important call. With the Asus ZenBook 14, you get the latest AMD Ryzen AI 5 430 processor, running on the Zen 5 architecture, along with 16 GB of LPDDR5X-7500 RAM and 512 GB PCIe 4.0 NVMe M.2 SSD. Right off the bat, the everyday performance of the ZenBook 14 is excellent. I write reviews, so my work mainly happens in Chrome, with about 20 tabs open for research. The laptop handled all that blissfully well, without a single hiccup. I could have multiple apps running in the background, too, and it wouldn’t break a sweat.

Sadly, I’m no coder or video editor, but I did try my hand at both. I downloaded VSCode and edited a 4K Instagram reel, both of which went surprisingly well. Sure, you cannot expect the laptop to handle ten 4K streams, but for casual editing and reels, it’s perfectly fine. Large coding projects are compiled in seconds, so if you’re a college student, this could be worth a look.

But what about the actual numbers? I know benchmarks don’t often tell the whole story, but they do help paint a picture. Keeping up with that spirit, I turned on performance mode and ran Cinebench R23, where the ZenBook 14 scored 1,098 in the single-core and 7,032 in the multi-core tests. In 3D Mark’s Wild Life Extreme test, the laptop scored 2,655 points. Finally, in the Night Raid test, the number reached 20,792.

Given the very decent benchmark scores, I thought we should play a few games on the ZenBook 14. But before that, please note that this is not a gaming laptop, by any stretch of the word. Still, if you only play eSports titles like Counter-Strike and Valorant, the ZenBook 14 might surprise you. At medium-to-high settings, I got over 100 FPS in both games, and the experience was jitter-free even during high-intensity matches.

Battery Life & Speakers

AMD has made quite a few gains in efficiency with its newest chips, and they help a lot with battery life on the Asus ZenBook 14. On a typical workday, which involves using Chrome, editing spreadsheets, and watching tons of YouTube videos, I get roughly 12 hours of SoT. This is almost MacBook-level battery life, and something I quite frankly didn’t expect. Charging is handled by a 65W fast charger, which gets the laptop from 20% to 80% in under an hour.

Usually, I’m not a fan of downward-facing speakers, since they inherently limit output depending on the surface’s hardness. While that factor is in play here, the speaker quality is top-notch. Compared to my aging MacBook Air, the ZenBook 14 sounds more full, with an emphasis on the mids. The vocals sound super clear, and the treble is nice. There’s Dolby Atmos support for the people who watch a lot of movies without headphones.

Verdict

At ₹114,990, the new Asus ZenBook 14 has everything a professional needs, and in my two weeks of testing, I couldn’t find anything that made me go, “Oh, I wish they had done this better.” The design is sophisticated yet stands out, with premium materials. The OLED panel is bright and colorful, and the touch functionality adds a new way of using the laptop, especially for kids. Performance keeps up with anything you might need, even demanding tasks, and a little dabble in the gaming world. Honestly, the ZenBook 14 gets a solid recommendation from me.

“Now I’m being taught how to do it the right way,” Ryan says. “And I’m being paid to do it.”

Ryan hopes Enhanced can pave the way for a separation of truly “clean” events and transparently juiced competitions. He even calls on events like the Olympics to increase testing strictness.

“What we’re doing is completely separate,” Ryan said during Friday’s media availability. “It’s marketing, it’s show business. And it should be separate.”

The financial argument and the idea of helping aging athletes prolong their careers make up the most compelling case I hear for Enhanced on its face—at least in terms of athletes’ motivations. But it’s the business side of the organization where some conflicts of interest become tougher to ignore.

Photograph: Etienne Laurent/Getty Images

Head to the products page of the Enhanced website and you’ll find what appears to be the organization’s spinoff of telehealth company Hims, but for PEDs. Products like copper peptides, sermorelin and testosterone injections are available alongside GLP-1s, semaglutide, and tadalafil.

Martin is open about the mission: To bring these products to the masses. He talks up required medical intake forms and regular check-ins with certified company doctors to avoid risks of mis- or over-use.

But if the Enhanced mission is successful, and PEDs become a bigger and more financially appealing part of sports, assuming these products will only be sought after and used in responsible ways is just as naïve as pretending doping hasn’t happened in the past at supposedly “clean” events. If anything, athletes in particularly disadvantaged financial situations might prioritize doping even more.

That’s where the dystopian feel bubbling below the surface becomes more palpable. There’s a distinctly MAHA undertone here, from investors like Thiel and Donald Trump Jr. to Enhanced Games founder Aron D’Souza describing RFK Jr. as “pro–human enhancement.”

While organizers wouldn’t give me any specifics on how much of Enhanced’s future will rely on product sales to fund prize pools and operations, it’s probably safe to assume investors like these will expect returns to remain involved.

It’s fair to wonder whether this is truly an attempt to remove stigmas and change sport. The overconfidence on display prior to the actual contests only drove home the feeling that this was more an elaborate money-making scheme than anything else.

So will the Enhanced Games usher in a new era of athletic capability and prowess? Maybe—at least if you can afford it.

Earlier this year, Lucra Sports founder and CEO Dylan Robbins did something that no one else has ever done.

He landed famed public investor Cathie Wood and her ARK Invest Venture Fund as a lead in a startup fundraising round.

Lucra announced last month that it raised a $20 million Series B, led by the ARK fund, with participation from several other VCs. Robbins attracted ARK even though the fund had previously gotten badly burned on a similar eSports company: Skillz, a skill-based gaming platform in which the fund invested heavily before divesting at a loss.

On top of that, Dylan landed this big fish as an investor even though his company is not in the one area that all VCs are currently chasing: AI.

Lucra offers white-label interactive gaming competitions as a novel kind of loyalty program for businesses that serve consumers. Rather than, say, earning points toward a coupon, Lucra’s clients offer online tournaments for prizes, or supports friendly wagers between their customers on who will win games. Its customers include Five Iron Golf, Dave & Buster’s, and Chess King.

Robbins told us there were two secrets in how he landed a big-name investor against such odds:

1. Be friendly to everyone, anywhere because you never know when a casual conversation will turn into your major investor.

2. Lead your pitch with AI even if you aren’t a famed AI scientist and aren’t building models, agents, or anything AI.

To the first point, the seeds to Lucra’s fundraising journey began when Robbins was playing darts in a New York bar. He met another guy at the dartboard, and they enjoyed a few games together.

“Six months later, we ran into each other at the bar again. The same darts bar. It’s like, ‘Good to see you. How’s it going?’ And we got to talking and I asked him what he did for work. And he told me he worked at ARK,” Robbins recalled.

Robbins told him about Lucra and the contact introduced him to the investment team at ARK, which wound up writing a small check in his Series A round.

“My first piece of advice on all of this is you never know who you’re talking to. Just go around, be nice, meet people, have fun,” Robbins says. Let that lead to good conversations, which will lead to introductions, he said.

Flash forward a few years to the end of 2025, when AI had overtaken venture funding like honeysuckle.

Lucra Sports had really found its lane with its white-labeling service. It was ready to raise a Series B to fuel growth and new ideas like adding mini-games into its offerings. (Lucra just invested in a mini-game development partner to build out this capability.)

But Robbins kept running into an AI-shaped wall.

“We were raising in Q4 of 2025, which was then, like even now, kind of peak AI mayhem,” Robbins said. “One out of every three calls, the first line, they would stop the meeting and say, oh, we’re only investing in AI now, I don’t want to waste your time. To the point where they wouldn’t even let me pitch.”

The rest told him they were only investing in AI after they heard the pitch.

So Robbins tried a new tactic. He adjusted his pitch and his deck to discuss AI right out of the gate. The revised pitch argued that if AI works, people are going to have more free time to play games with friends at the bar or online — hence his business will be a winner — and if it doesn’t, a non-AI bet starts to look like smart diversification. It was a hedge either way.

“It was a small cohort of people that would really take it seriously,” he said of his pitch. ARK, fortunately, was one of them. Once committed, the lead investor made introductions to other VCs to help fill out the round.

Underpinning all of this were good business fundamentals, including “consistent year over year growth, not just one spurt,” he said.

The final lesson Robbins learned was that, especially for a non-AI business, VCs want to hear a big dream. Robbins had one: a total addressable market of anyone who plays games of any kind, from pickleball to Wordle.

“So our TAM is almost every American that’s 18 to 70, right?” Robbins said. Even so, he had one VC send a rejection that he printed out and posted to the wall.

“I sent them our growth chart and our TAM, which was like crazy, up into the right growth potential, huge, big, billions of TAM. And the response was: ‘TAM’s too small.’ That was the response. Like, our growth rate was too slow,” he said.

He said this was a “reminder” to him “to think even bigger.”

“I have to put myself in that mindset and really swing for the fences if I want to raise venture capital money,” he added.