Do you find yourself typing the same things over and over?
If the answer is yes — and if you type enough text during the workweek, it almost certainly will be — I’d highly recommend you install a text expansion application. These text shortcut tools are a must-have productivity upgrade for many knowledge workers, and they can even help you at home. If you haven’t given one a go yet, you’re missing out.
There are am wide variety of text expanders to choose from. Many are paid tools intended for businesses, complete with pricey subscriptions and features that only make sense for large teams and complex professional workflows.
(These types of apps are particularly relevant in the PC arena; text shortcut tricks are built into Android phones and Apple devices, but on Windows, you need a third-party application to take advantage of them.)
I’ll cover the text expanders I recommend — easy-to-use, free solutions for Windows PCs. If your workplace pays for a different one, that’s great. Otherwise, I suggest starting with one of the options here.
Up first is Beeftext, a completely free, open-source, and as-simple-as-it-gets text expansion tool for Windows PCs. My colleague Jared Newman recommended it at PCWorld — and for good reason.
After installing and launching Beeftext, you’ll see a little red “Bt” icon in your system tray. Open it to get started.
In Beeftext, combos are text expansion shortcuts. You’ll click Combos > New to create a new one. Then, you’ll just need to enter a Snippet and a Keyword. When you type the keyword in any application, Beeftext will replace it with the snippet.
For example, you might want to try creating the keyword /add and specifying your address. Then, you can type that keyword in any application on your PC to quickly insert your address.
While creating a combo, you just need to fill in the “Snippet” and “Keyword” boxes.
Chris Hoffman, IDG
It’s really that simple. You could enter /# and associate that with your phone number. You don’t have to use a slash — it’s just an easy way to create memorable shortcuts that you won’t type accidentally.
You can do a lot with this type of tool. If you often find yourself responding to a customer, client, or even just a friend with a specific message, you might create a snippet for that message so you can quickly type it in emails. If you often end up inserting some kind of signature or disclaimer message, you might create a snippet for that. It’s really up to you.
Autocorrect: Built right into Office apps
It’s worth noting that Office apps like Microsoft Word, Excel, OneNote, and Outlook have a text-expansion-style feature built right into them. It will only work inside those Office apps. But, if you only need this feature while working in those specific applications, it’s perfect. And it might be the only solution for workers who can’t install anything else on their PCs.
First, open the application you want to use — such as Microsoft Word or OneNote. Click the “File” menu and select “Options.” Click the “Proofing” option in the left sidebar, and then click the “AutoCorrect Options” button.
The AutoCorrect window will appear. Under “Replace text as you type,” add something you want to replace at the left and the text you want to replace it with at the right. Then, click the “Add” button to add it to the list, and click “OK.”
For example, as in the above example, you could enter /add and have Office replace that text with your address when you type it.
AutoCorrect is incredibly convenient. I wish it worked elsewhere in Windows, too!
Chris Hoffman, IDG
Want to use formatted text — or text with multiple lines? Type some text in your document, and then select it with your mouse. Then, click File > Options > Proofing > AutoCorrect Options. The text you selected will be prefilled there, and you can associate a text shortcut with it.
Text Blaze: An extension for Chrome and Edge
Last but not least, Text Blaze is another exceptional option for Windows-based text replacement. (We use this application at our newsletter-focused small business The Intelligence.) While it’s not completely free, it does offer a free plan for individuals. That plan gives you up to 20 snippets, which is plenty for many people.
Text Blaze lets you specify a “shortcut” that will be replaced by a “snippet” when you type It on any web page in your browser.
Chris Hoffman, IDG
On the business front, the tool has some useful features that work well for our team, such as sharing snippets with other people and dynamically entering information in snippets. That’s more than many users will need, though. The paid subscription won’t be necessary if you just want some quick text shortcuts in Google Chrome for your own use.
Text Blaze does have a Windows application, too — but, in my experience, it works better and faster as a Chrome extension.
More text expansion options for Windows PCs
With all the text expansion tools available, it’s hard to narrow the list. If you do some web searches, you’ll see a lot of other similar options. Here are a few other popular possibilities and what you need to know about them — including why they didn’t earn one of my top recommendations:
Espanso is a completely free and open-source text expander that works on Windows as well as other platforms. However, it doesn’t have a convenient graphical interface, so you’ll have to write out your “matches” and “replaces” in a text editor. If that works for you, go right ahead — it will be the tool of choice for many geeks.
TextExpander is incredibly polished and feature-filled. But this tool doesn’t have a free version. While it might be a great fit for some businesses who want to use the sharing functionality with their team, the $40/year subscription here isn’t a great fit for the average individual.
Phrase Express is another application with a long history. But, starting at $100 per license, it’s probably not the ideal choice for most people.
I can’t emphasize enough: There are many other text expansion tools that’ll work with Windows. Lots of them likely work well, too, but most aren’t free or have other issues that make them not-entirely-optimal choices for most common purposes.
Bonus: Pin text to Clipboard History in Windows
Don’t want to install a text expander? You can also “pin” items to your clipboard history in either Windows 11 or Windows 10. To open it, press Windows+V and choose to enable clipboard history if you haven’t already done so.
Then, after copying some text or an image to your clipboard, you can press Windows+V and use the “pin” option to save it to your clipboard history for later. You can then press Windows+V in the future and select a pinned item to quickly paste it.
The Clipboard History tool is useful for both text and images.
Chris Hoffman, IDG
It’s not a text expander, exactly — but it is a way to quickly insert some saved text. And it’s that end result that counts the most.
I do. It was March 10, 2000. The NASDAQ Composite index peaked at 5,048.62 points. That doesn’t sound like much today, but it was up 800% from 1995. Then people started to get nervous about the true value of the dot.com businesses driving the market, interest rates started to rise, and by October 2002, the NASDAQ had fallen 78% from its peak.
We’re not even a month into 2025, and the new Chinese AI program DeepSeek this week sparked Nvidia’s $465 billion rout, the biggest single-day drop in US stock market history.
This is just the beginning. It’s not that DeepSeek is so much better than ChatGPT’s OpenAI or any of the other popular generative AI (genAI) tools. Maybe it is, maybe it isn’t. (Just don’t ask it about what happened at Tiananmen Square in 1989.)
No, what matters is that DeepSeek requires an order-of-magnitude less computing power to achieve similar results. By programmer Simon Willison’s count, “DeepSeek v3 trained on 2,788,000 H800 GPU hours at an estimated cost of $5,576,000. For comparison, Meta AI’s Llama 3.1 405B (fewer than DeepSeek v3’s 685B parameters) trained on 11x that; it required more than 30.8 million GPU hours and was trained on 15 trillion tokens of data.
In other words, “large language models (LLM) pricing is going to collapse.
“The scary part is the LLM giants didn’t have profit margins to begin with,” he continued. “LLMs are going to commodity in a hurry.” That’s bad news for companies such as OpenAI. Sure, the company’s market value stands at $157 billion — but it’s still losing billions of dollars. If OpenAI’s customers decide they don’t want to pay its rates when they can get much the same service for a fraction of the cost from DeepSeek, where does that leave OpenAI?
What happens then? Look at it this way: today’s overheated stock market is driven by the Magnificent Seven— companies that with one exception (Tesla) have one thing in common: they’re all heavily invested in AI hype. Even Tesla is heavily invested in what Elon Musk calls real-world AI.
It’s not just the Magnificent Seven; pick-and-shovel AI companies such as Oracle, Super Micro, and Nebius (formerly Yandex) are also in a world of trouble.
Bratin Saha, DigitalOcean’s chief product and technology officer, sees the potential for a relatively small company to become a big player. “DeepSeek AI is the ‘Android moment for AI’ (given that ChatGPT was likened to the ‘iPhone moment for AI’), as it shows groundbreaking AI products from the open-source community,” Saha told me.
“DeepSeek AI democratizes AI and cloud computing because it shows you do not need multi-billion dollar investments for compelling innovation,” he said. “It lowers the barrier for small and medium enterprises and individual developers to work with AI.”
Exactly.
The way Stephen O’Grady, co-founder and industry analyst with RedMonk, sees it, enterprises have two major AI concerns. How trustworthy the technology is and its cost. As O’Grady points out, “Enterprises have been shocked, in many cases, at the unexpected costs – and unclear returns – from some scale investments in AI.”
What if enterprises don’t have to rely on closed, private models for leading-edge capabilities?
What if training costs could be reduced by an order of magnitude or more?
What if they don’t require expensive, state-of-the-art hardware to run their models?
This changes everything for businesses that want to use AI and multi-billion dollar companies that want to sell AI to you. It’s a sea change.
“What the markets reacted to was DeepSeek’s ability to build a model that rivaled OpenAI’s sophisticated o1 reasoning model and Anthropic’s Sonnet model for pennies on the dollar on a fraction of the compute,” Jim Zemlin, executive director of the Linux Foundation, wrote. “It also appears that DeepSeek did this using well-known techniques. There were no massive algorithmic breakthroughs, just very clever engineering.”
Still,”a small team in China took a fresh look at a problem and came up with a novel approach that reduced the cost of chain-of-thought reasoning by 50x (if DeepSeek’s postings are accurate) and then published a paper fully describing their process, allowing the community to benefit from their learnings,” Zemlin said. “We need MORE of this progress, not less.… This is a struggle over open markets between the forces of open and the forces of closed.”
That may be good for open source and for AI, but for our stock market, which is largely driven by AI, it’s another matter. The Magnificent Seven accounted for over 50% of the S&P 500’s gains in 2024. If they collapse, a good chunk of the market will follow.
I’m not normally a bear, but I’ve seen this rodeo before. Oh, AI will survive, and eventually, it will prove useful and profitable. But, in the meantime, I can see an AI crash coming, and it won’t be pretty.
It’s fascinating to see how few people actively use some of Android’s most useful features.
I had such a revelation around the platform’s long-standing split-screen option this month. With Google working to expand Android’s screen-splitting system and make it even more flexible with an upcoming Android version, I decided to ask the smart and savvy readers of my Android Intelligence newsletter how often they actually tap into that Android-exclusive advantage.
And the survey saaaaaaaaaaays: A staggering 57% of folks rarely to never rely on Android’s split-screen option, while 19% didn’t even know such a thing existed. Just 6% said they use it once or twice a week, meanwhile, and a meager 5% almost every day.
But my goodness, can it ever be a handy efficiency-enhancer to wield. And beyond the basic act of splitting your screen in half and interacting with two apps at the same time, Android’s split-screen system holds some splendidly advanced options for pushing your productivity to powerful new heights.
My friend and fellow tech-sorcery adorer, prepare to be enlightened.
[Psst: Want even more advanced Android knowledge? Check out my free Android Shortcut Supercourse next to learn tons of time-saving tricks for your phone.]
Android split-screen trick #1: The split slide
Up first is an easy one to wrap your mind around. Once you’ve got two apps open together on your favorite Android device — and if you don’t know how to do that yet, pause for a moment and start here for the 101-level steps — take note: You can slide your finger up and down (or side to side, if your device is in a landscape orientation) on the line between the two sides of your screen to change the spacing and see more of either app in the split.
srcset="https://b2b-contenthub.com/wp-content/uploads/2025/01/android-split-screen-adjust-size.webp?quality=50&strip=all 900w, https://b2b-contenthub.com/wp-content/uploads/2025/01/android-split-screen-adjust-size.webp?resize=292%2C300&quality=50&strip=all 292w, https://b2b-contenthub.com/wp-content/uploads/2025/01/android-split-screen-adjust-size.webp?resize=768%2C788&quality=50&strip=all 768w, https://b2b-contenthub.com/wp-content/uploads/2025/01/android-split-screen-adjust-size.webp?resize=679%2C697&quality=50&strip=all 679w, https://b2b-contenthub.com/wp-content/uploads/2025/01/android-split-screen-adjust-size.webp?resize=164%2C168&quality=50&strip=all 164w, https://b2b-contenthub.com/wp-content/uploads/2025/01/android-split-screen-adjust-size.webp?resize=82%2C84&quality=50&strip=all 82w, https://b2b-contenthub.com/wp-content/uploads/2025/01/android-split-screen-adjust-size.webp?resize=468%2C480&quality=50&strip=all 468w, https://b2b-contenthub.com/wp-content/uploads/2025/01/android-split-screen-adjust-size.webp?resize=351%2C360&quality=50&strip=all 351w, https://b2b-contenthub.com/wp-content/uploads/2025/01/android-split-screen-adjust-size.webp?resize=244%2C250&quality=50&strip=all 244w" width="900" height="924" sizes="(max-width: 900px) 100vw, 900px">See more, see less: With a simple slide in Android’s split-screen mode, you can adjust as you see fit.
JR Raphael, IDG
If you slide your finger all the way down (or to the side), meanwhile, you’ll exit Android’s split-screen mode entirely and go back to seeing just a single app at once.
Android split-screen trick #2: The tab grab
Typically, when you’re starting up a new split-screen, you select two different apps from your recently used app list to get things going.
But — little-known fact! — you can also select two different tabs from the Android Chrome browser and then view and even scroll through ’em side by side simultaneously.
This one’s a little tricky to find, but once you know the trick, it’s actually quite easy:
First, open up Chrome.
Then, fire up a split-screen between that and any other app, using the standard Android split-screen shortcut. It doesn’t matter what other app you choose, as we’ll override it a moment.
Now, in the Chrome half of your screen, tap the three-dot icon in the upper-right corner and select “New window.”
Alternatively, you can open any web page in the initial Chrome view and press and hold your finger onto any link within that page — then, in the menu that pops up, find and select the “Open in other window” option.
And hey, how ’bout that?!
Two Chrome apps side by side in a split — what’ll they think of next?!
JR Raphael, IDG
You’ve now got two Chrome tabs together on your screen, and you can change ’em or interact with ’em in any you want. You can even move tabs from one side of the screen to the other, if you’re really feeling saucy.
Android split-screen trick #3: The email divide
Similar to Chrome, Google’s official Gmail app for Android has its own crafty way to create a split-screen using multiple views from a single inbox.
This one works a wee bit differently, and it’s even easier to pull off:
That’ll put your new message draft on the first half of the screen. And you should then see your Gmail inbox view as an available option for the other half — or, if you’re on a Samsung device, be able to find and select Gmail from the list of other choices to get to that same spot.
Email in multiple panels, thanks to Android’s split-screen system.
JR Raphael, IDG
An active draft and your inbox or any other email on-screen together — doesn’t get much more convenient than that.
Android split-screen trick #4: The split-screen redo
In more recent Android versions, anytime you’ve created a split-screen between two apps, you can then find and return to that same exact setup — with both those same apps in the same split-screen position — even after you’ve moved away and shifted your focus elsewhere.
Just look in your recent apps overview — y’know, the thing you get to by swiping up about an inch from the bottom of the screen and stopping, with Android gestures, or by tapping the either square-shaped or three-vertical-line icon at the bottom of your screen, if you’re still clingin’ to the old legacy three-button nav approach — and you should see any recently accessed splits there and ready to return to with a couple quick taps.
srcset="https://b2b-contenthub.com/wp-content/uploads/2025/01/android-split-screen-recents.webp?quality=50&strip=all 900w, https://b2b-contenthub.com/wp-content/uploads/2025/01/android-split-screen-recents.webp?resize=292%2C300&quality=50&strip=all 292w, https://b2b-contenthub.com/wp-content/uploads/2025/01/android-split-screen-recents.webp?resize=768%2C788&quality=50&strip=all 768w, https://b2b-contenthub.com/wp-content/uploads/2025/01/android-split-screen-recents.webp?resize=679%2C697&quality=50&strip=all 679w, https://b2b-contenthub.com/wp-content/uploads/2025/01/android-split-screen-recents.webp?resize=164%2C168&quality=50&strip=all 164w, https://b2b-contenthub.com/wp-content/uploads/2025/01/android-split-screen-recents.webp?resize=82%2C84&quality=50&strip=all 82w, https://b2b-contenthub.com/wp-content/uploads/2025/01/android-split-screen-recents.webp?resize=468%2C480&quality=50&strip=all 468w, https://b2b-contenthub.com/wp-content/uploads/2025/01/android-split-screen-recents.webp?resize=351%2C360&quality=50&strip=all 351w, https://b2b-contenthub.com/wp-content/uploads/2025/01/android-split-screen-recents.webp?resize=244%2C250&quality=50&strip=all 244w" width="900" height="924" sizes="(max-width: 900px) 100vw, 900px">You can find recent screen-splits in the Overview area on recent Android versions.
JR Raphael, IDG
Android split-screen trick #5: The instant split
You don’t have to rely on your Android screen-splitting history to get back to a favorite app split in a jiff. A clever little app called Be Nice makes it as easy as can be to create custom one-tap shortcuts for starting up specific pairs of apps in a split on demand — just like opening any regular ol’ individual app on your device.
If you’ve got a large-screen Android device — be it a tablet or an expanded foldable phone — with Android 15 in place, you can actually accomplish something similar without any third-party add-ons.
Speaking of tablets and foldables, if you’re rockin’ such a plus-sized device and you’ve got a reasonably recent Android version on it, you can spin yourself into a split in a split-second simply by summoning the taskbar in the Android large-screen environment.
Just swipe up gently from the bottom of your screen to get the taskbar to appear, then drag an icon from that area up into your main screen space.
See?
srcset="https://b2b-contenthub.com/wp-content/uploads/2024/09/google-pixel-9-pro-fold-taskbar.webp?quality=50&strip=all 600w, https://b2b-contenthub.com/wp-content/uploads/2024/09/google-pixel-9-pro-fold-taskbar.webp?resize=289%2C300&quality=50&strip=all 289w, https://b2b-contenthub.com/wp-content/uploads/2024/09/google-pixel-9-pro-fold-taskbar.webp?resize=162%2C168&quality=50&strip=all 162w, https://b2b-contenthub.com/wp-content/uploads/2024/09/google-pixel-9-pro-fold-taskbar.webp?resize=81%2C84&quality=50&strip=all 81w, https://b2b-contenthub.com/wp-content/uploads/2024/09/google-pixel-9-pro-fold-taskbar.webp?resize=463%2C480&quality=50&strip=all 463w, https://b2b-contenthub.com/wp-content/uploads/2024/09/google-pixel-9-pro-fold-taskbar.webp?resize=347%2C360&quality=50&strip=all 347w, https://b2b-contenthub.com/wp-content/uploads/2024/09/google-pixel-9-pro-fold-taskbar.webp?resize=241%2C250&quality=50&strip=all 241w" width="600" height="622" sizes="(max-width: 600px) 100vw, 600px">Splitting your screen the simple way with the taskbar on a large-screen Android device.
JR Raphael, IDG
Whee!
Android split-screen trick #8: The Keep 2-in-1 special
Also specific to large-screen Android environments is a nifty little trick that lets you create a split-screen view right within the Google Keep Android app — without having to rely on the standard system-level screen-splitting feature at all.
Just open up any note inside the app, and as long as your screen is wide enough, you’ll see the note show up alongside your main note list. And then, from there, you can press and hold your finger onto the line separating the app’s two panels — the note list and whatever individual note you’re actively viewing — and slide your finger in either direction to change the panels’ sizes.
From Keep to calendar, the Google Calendar Android app has a similarly sneaky bit o’ screen-splitting sorcery built right in for larger Android devices.
In Calendar, you can trigger a split by tapping on a specific individual event from within a calendar view. That’ll cause your screen to divide and the event to appear alongside the full calendar interface — again, on a larger-screened Android gizmo only — and just like with Keep, you can then press and slide your precious fingie along the line between those panels to adjust each side’s size.
Pretty handy — and we’ve got two easily overlooked Android split-screen tricks left yet…
Android split-screen trick #10: The floating web window
In addition to the snazzy Chrome tab screen-splitting tricks from a second ago, you can also pop any web page out into a floating window that then sits atop anything else you’re doing on your device — like a more versatile sort of split-screen setup, in a sense.
This one requires a helping hand from a third-party app called, fittingly enough, Float Browser.
Last but not least is a final native Android feature for splitting your screen in an unconventional but effective way — at least, for certain specific scenarios.
It’s Android’s picture-in-picture system, which makes it easy to keep an actively playing video or even a Google Maps navigation visible on your screen while you also go about other tasks.
In apps that support the function, you can start a picture-in-picture view simply by heading back to your home screen while a video or navigation is playing (though some apps, like YouTube, do have certain restrictions in place for when the feature can be used).
And there ya have it: 11 excellent ways to make the most of Android’s split-screen framework. Use your newfound power wisely — and as with any splits, be sure to do plenty of gentle stretching.
↪️ Up next: Get six full days of advanced Android knowledge with my free Android Shortcut Supercourse. You’ll learn tons of time-saving tricks for your phone!
President Donald Trump has suggested plans to impose tariffs of up to 100% on semiconductors imported from Taiwan, aiming to boost US manufacturing but risking higher costs and supply chain disruptions for tech firms and enterprises.
“In the very near future, we’re going to be placing tariffs on foreign production of computer chips, semiconductors, and pharmaceuticals to return production of these essential goods to the United States of America,” Trump said during a recent address.
AWS and Microsoft could potentially face an investigation of their cloud businesses in Britain should the Competition and Markets Authority (CMA) follow the recommendation of an independent inquiry group that reports into the UK’s antitrust regulator.
In a release issued on Monday, the group asks that the CMA use its powers under the Digital Markets, Competition and Consumers Act 2024 (DMCCA), to “designate the two largest providers, AWS and Microsoft, with strategic market status (SMS) in relation to their respective digital activities in cloud services.”
The new legislation took effect at the start of the year, and UK law firm Macfarlanes wrote in an advisory on the new legislation, issued on January 13, that the “UK’s answer to the EU’s Digital Markets Act (DMA) empowers the CMA to designate firms as having “Strategic Market Status” (SMS) in relation to a particular ‘digital activity.’ Once designated, the CMA can impose ‘conduct requirements’ (CRs) on those firms or introduce ‘pro-competition interventions’ (PCIs) to achieve positive outcomes for UK consumers and businesses.”
There is a special Copilot key on some Windows laptops that can be used to launch Microsoft’s AI assistant — and now support for the key is coming to Linux.
That support has been added in version 6.14 of Linux, though exactly what it will be used for depends on which Linux-based operating system you are running. Most users are likely to use the key to open any generative AI assistant.
Version 6.14 offers several other new features, including expanded support for hand controllers, according to Phoronix.
They may be stressed about shadow IT, security, endpoint management, and the recent proliferation of artificial intelligence (AI) tools. They’re challenged not just by static budgets and rising costs, but by the proliferation of tools they now must use to do their job. And while IT admins are not exactly struggling for work, there is plenty for them to do.
Those are some of the findings in JumpCloud’s latest survey of IT admin decision-makers across the US, UK, and Australia. The report confirms that the workplace continues to become increasingly multi-platform, with 27% of enterprise employees now preferring to use Macs, up from low-single figures at the turn of the century.
“Windows use has shown the most significant decrease over the last six months, compared to macOS and Linux, which both increased,” said JumpCloud.
When asked about the breakdown of their organization’s device type, admins reported Windows devices comprise 56% (down from 63% in Q3 2024), macOS 27% (up from 24% in Q3 2024), and Linux 20% (up from 18% in Q3 2024).
Mac adoption in the enterprise is certain to continue to climb. The tech support cost overhead of managing Windows systems means a migration might represent low-hanging fruit for many enterprise leaders working to squeeze more from their budgets. Initial cost aside (and the difference between Mac and equivalent Windows systems is smaller now than ever), the total cost of ownership has a significant impact on budgets.
Make no mistake. Budget-wrangling really is a thing: 39% spend up to half of their entire budget on licensing fees. While this reflects the grim reality that tech providers of all kinds are forcing subscriptions onto their customers, it represents a massive increase in such costs. In Q3, 2024 just 28% of admins endured similar budget erosion for licensing fees.
What do you get for your money?
Windows and other Microsoft devices were seen as the most difficult things to manage by 23% of respondents. To be fair, Apple devices were seen as difficult to manage by 19% of IT admins, with Linux winning unspoken praise — just 14% of admins saw it as the most difficult.
All the same, what that difference in Windows-vs-Mac management difficulty means is that the Windows experience is more abrasive, which — in conjunction with the upcoming Windows licensing replacement cycle — means IT will be tempted to look at alternatives.
Perhaps that’s why 43% of admins expect macOS device use to increase in the coming year, though 54% also anticipate increased use of Windows. The default rate — enterprises dropping support for either platform — is fairly equal, though Apple has the edge.
On-prem, off-prem, and multicloud
Admins are also frustrated at the complexity of managing cloud and multi-cloud setups. That’s turning into a big opportunity for managed service providers (MSPs) who increasingly offer to ease the pain of managing multi-cloud setups. MSPs aren’t just about cloud services management, of course. But it does appear to be their time to shine, with 93% of organizations already using or considering an MSP. Their role also seems to be changing, as they’re increasingly seen as trusted advisors.
Can AI supplement IT roles? Admins see the technology as both a risk and an opportunity. But organizations appear to be accelerating AI deployments, with 15% of admins warning the tech is being put in place too fast and 67% believing these deployments are outpacing organizational ability to protect against threats.
“Keeping pace with all the improvements and changes keeps me up at night,” one anonymous survey respondent told JumpCloud. “AI has bought a new way of doing business and requires major adjustments.”
Back to the (AI) future
It’s not just deployment that has admins spooked. Thirty-seven percent of them fear AI will take their jobs — and 56% of corporate vice presidents now worry AI will replace them, up from 29% a year ago. All the same, fear of redundancy is endemic across every role.
How is AI being managed? Most enterprises are taking steps to accommodate its use, with just 21% having taken no steps or put AI restrictions in position. Almost half (49%) of companies have developed policies to guide employee use of AI, with 47% encouraging use of tools such as ChatGPT.
With data being the new gold, it’s no surprise that 28% of IT admins said their companies now have controls in place to prevent employee use of AI. “To harness [AI’s] power responsibly, organizations must lead with clear governance and innovation frameworks that balance opportunity with risk,” JumpCloud said.
All the same, unauthorized use of AI continues, and just like any other form of Shadow IT this proliferation is a big problem for IT.
The usual suspects
The number of admins concerned about the use of apps and devices that aren’t managed has increased again, with 88% of IT admins now worried about this. They estimate that most employees use between one and five unauthorized applications.
There are lots of reasons for this, one of them being the speed at which businesses are moving, which means the current needs aren’t being met, driving employees to seek solutions that fit. And while you’d expect IT to spend time handling this, lack of time and lack of visibility into all the apps employees use means that just as fiscal budgets demand careful juggling, so too does precious IT time.
What else is eating that time? Security. It currently consumes the lion’s share of IT budgets. A plurality of organizations (47%) spend between 10% and 25% of their yearly IT budget on cybersecurity; another 24% spend 26% to 50%; 5% spend more than half their budget on security; and 24% less than 9% on security. In other words, security remains a tidy little earner for vendors, and a significant revenue expenditure line item for IT.
Almost half (46%) of organizations report that they have fallen victim to a cyberattack. AI-augmented attacks are also proliferating — this is now the third-biggest security concern after phishing and shadow IT. Man-in-the-middle attacks, MFA hacks and security breaches in partner organizations are also on the rise.
In other words, security is an endless feast of fear for some, and of revenue for others. Of course, things might be better if there were platform choices that could mitigate this attack surface.
Chinese start-up DeepSeek rocked the tech industry Monday after the company’s new generative AI (genAI) bot hit Apple’s App Store and Google Play Store and downloads almost immediately exceeded those of OpenAI’s ChatGPT. US AI model and chipmaker stock prices were hit hard by the newcomer’s arrival; Google, Meta and OpenAI all initially suffered and chipmaker Nvidia’s stock closed the day down 17%. (The tech heavy Nasdaq exchange lost more than 600 points.)
DeepSeek’s open-source AI model’s impact lies in matching US models’ performance at a fraction of the cost by using compute and memory resources more efficiently. But industry analysts believe investor reaction to DeepSeek’s impact on US tech firms and others is being dramatically exaggerated.
“The market is incorrectly presuming this as a zero-sum game,” said Chirag Dekate, a vice president analyst at Gartner Research. “They’re basically saying, ‘Maybe we don’t need to build data centers anymore, maybe we’re not as energy starved because DeepSeek showed us we can do more with less.’”
Giuseppe Sette, president of AI tech firm Reflexivity agreed, stressing that DeepSeek took the market by storm by doing more with less.
“In layman terms, they activate only the most relevant portions of their model for each query, and that saves money and computation power,” Sette said. “This shows that with AI, the surprises will keep on coming in the next few years. And even though that might be a bit of a shocker today, it’s extremely bullish in the long-term — because it opens the way for deeper and broader adoption of AI at all scales.”
In essence, the markets have overlooked that companies such as Google, Meta, and OpenAI can replicate DeepSeek’s efficiencies with more mature, scalable AI models that offer better security and privacy.
“This is not a ‘the sky is falling moment’ for markets. I think they should take a close look at what this actually is: there are techniques you can implement to more effectively scale your AI models,” Dekate said.
Another looming problem for the newcomer is that DeepSeek is purported to filter out content that could be viewed as critical of the Chinese Communist government. DeepSeek’s release of its R1 and R1-Zero reasoning models on Jan. 20 quickly drew attention for two key aspects:
DeepSeek requires less memory and compute power, needing fewer GPUs to perform the same tasks as other models.
DeepSeek claims its breakthroughs in AI efficiency cost less than $6 million and took less than two months to develop.
John Belton, a portfolio manager at Gabelli Funds, an asset management firm whose funds include shares of Nvidia, Microsoft, Amazon, and others, said DeepSeek’s achievements are real, but some of the company’s claims are misleading.
“No, you cannot recreate DeepSeek with $6 million and the extent to which they distilled existing models (took shortcuts potentially without license) is an unknown,” Belton said via email to Computerworld. “However, they have made key breakthroughs that show how to reduce training and inference costs.”
Belton also pointed out that DeepSeek isn’t new. Its creator, Liang Wenfeng, a hedge fund manager and AI enthusiast, published a paper on the performance breakthroughs more than a month ago and released a model with similar methods a year ago.
DeepSeek improved memory bandwidth efficiency with two key innovations: using a lower-position memory algorithm and switching from FP32 (32-bit) to FP8 (8-bit) for model precision training. “They’re using the same amount of memory to store and move more data,” Dekate said.
One analogy would be to consider the onramp to a major city highway — the highway being the data path. If the onramp only has one lane, there are only two ways to address traffic congestion:
Increase the width of the roadway to fit more traffic
Reduce the size of the vehicles to more fit on the roadway
DeepSeek did both. It created smaller vehicles, i.e., it used smaller data packets (8-bit) and therefore was able to pack more data into the same footprint.
The second key innovation was optimizing and compressing the key-value cache. DeepSeek used compression algorithms to reduce memory by processing prompts in two phases: decomposing and generating responses, both relying on efficient key-value cache use.
“They utilized underlying compute and memory resources incredibly efficiently,” Dekate said. “That is an amazing accomplishment, because they’re utilizing the underlying GPU resources more productively. Their models are able to perform at leadership-class levels while using a relatively lower scale of resources.”
Enterprises can benefit as well by adopting the techniques introduced by DeepSeek because it reduces the cost of adoption by using fewer compute resources for inferencing and training. Lower model costs should benefit innovators such as OpenAI and reduce the cost of applying AI across industries.
By using resources more efficiently, DeepSeek enables faster, broader AI adoption by other companies, driving growth in AI development, demand, and infrastructure.
And in the end, DeepSeek’s algorithm still needs AI accelerator technology to work — meaning GPUs and ASICs.
“It’s not the case that DeepSeek just woke up one day and had an amazing breakthrough. No, they’re using sound engineering techniques and they’re using some of the leading AI accelerators — and GPUs happen to be table stakes,” Dekate said. “And they use thousands of them. It’s not like they discovered a new technique that blew this whole space wide open. No. You still need AI accelerators to perform model training.”
Even in the most pessimistic view, if AI costs drop to 5% of those from other leading AI models, that efficiency eventually benefits those other models by reducing their costs, allowing for faster model adoption.
For enterprises, Dekate said, it’s worth exploring DeepSeek and similar models internally and in private settings. “Your legal team evaluates the terms and conditions of your ecosystem quite extensively. They’ll ask if privacy is protected. Are the data sources filtered? Are AI model responses filtered in any sense?” he said.
Before jumping in, enterprises should carefully consider these details. “Models like Gemini and GPT offer reliable, secure responses with enterprise-level protections, unlike many open models that lack these controls,” Dekate argued.
“Once everything settles, the net-net is that DeepSeek has developed very specific capabilities that are quantitative and that’s something to learn from, just as they did from Llama 3,” Dekate said.
Chinese AI startup DeepSeek said it was hit by a cyberattack, prompting the company to restrict user registrations and manage website outages as demand for its AI assistant soared.
According to the company’s status page, DeepSeek has been investigating the issue since late evening Beijing time on Monday.
Chinese AI startup DeepSeek said it was hit by a cyberattack, prompting the company to restrict user registrations and manage website outages as demand for its AI assistant soared.
According to the company’s status page, DeepSeek has been investigating the issue since late evening Beijing time on Monday.
