Adobe Acrobat users can now generate and edit images in PDFs via text prompts with the addition of Adobe’s Firefly AI image model.

The update gives Acrobat and Reader users access to some of the same image creation and editing capabilities available in Adobe design apps such as Photoshop, without the need to switch tools. This means it’s possible to create new images from scratch using Adobe’s latest Firefly Image 3 model — which Adobe claims will enable greater realism and more control over the images produced compared to previous models — or manipulate existing content with Generative Fill and Remove Background tools.

The Firefly image feature, announced Monday, is available for Adobe Acrobat Standard and Pro customers; they will receive 250 “generative credits” to use with the Edit Image and Generate Image features each month at no additional cost through January 2025 an Adobe spokesperson said. The price after that date point has yet to be determined. 

“Adobe has done workers a service by building technologies like the generative capabilities of Adobe Firefly and working to integrate them across a range of their applications and tools,” said Matt Arcaro, IDC research director for computer vision and AI. By building Firefly into familiar apps such as Acrobat, the level of friction involved in accessing a new technology is reduced, he said. 

Adobe has also added new features to the AI Assistant for Acrobat and Reader apps announced in February. 

The generative AI (genAI) assistant can now summarize information from multiple documents — including PDF, Word, PowerPoint, and other file formats — at the same time. That lets users identify common trends and generate an overview summary with citations that point to the source of information, Adobe said. Users can then direct the AI assistant to format the content for a specific purpose, such as turning it into an email that can be sent to colleagues. 

The ability to query multiple documents adds value to Adobe’s “document chat” tool, said Arcaro. “This creates a significant speed up for gathering key insights, but also to put together a more comprehensive perspective across multiple similar or different documents,” he said.

The Acrobat AI assistant can also be used to summarize the transcription of a Microsoft Teams meeting or webinar, or upload a “VTT” transcript file generated in another meeting app. 

The Acrobat AI Assistant is available for free until June 28, after which it will cost an additional $4.99 each month for individual Acrobat Standard and Pro subscribers under Adobe’s “early access pricing.” This pricing is available up to Sept. 4. 

Adobe didn’t respond to a request for pricing details after this date. 

If you have an Apple ID, you can use iCloud with Windows, which means you can access your email, videos, photos, calendars, bookmarks, files and other iCloud data from your PC – useful when you’re at your desk or traveling without an Apple device, or if you are inside Boot Camp on your Mac.

How to use iCloud with Windows

There are two ways to use iCloud with Windows: via a browser or using a dedicated application called iCloud for Windows, but there are differences between the two:

The browser

On the surface, the most straightforward approach to using iCloud with Windows systems might appear to be accessing your iCloud account from your favorite web browser.

The app

That’s great up to a point, but to unlock the power of sync across all the devices you have logged into your iCloud account, including your Windows systems, it is far better to download and install Apple’s iCloud for Windows software on your PC. The software is available for free download directly from Microsoft’s app store.

Once installed and set up, you’ll be able to use iCloud to get to all your photos, documents and bookmarks on your Windows PC to update automatically across all your Apple devices, and vice versa, which is what you get when you use iCloud for Windows. 

Borrowed from One Drive

The most recently published version of iCloud for Windows uses Microsoft’s Cloud Files API, (One Drive’s Files On Demand feature) which makes the solution more robust and faster.

This means new assets will swiftly sync across all your devices, “enabling users to be more productive offline on mobile devices and quickly share files on iOS,” according to Microsoft Engineering General Manager Giorgio Sardo. In part, this is because the system is smart enough to just download a “breadcrumb” (around 1k in size) of each of your files, downloading the full item on request.

How does iCloud Drive work?

If you already use iCloud Drive, you’ll be familiar with what it does. If not, then the easiest way to understand the feature is as online storage for all the files and projects you are working on. Anything you save into any folder of the iCloud Drive will sync across all your Apple ID devices (it’s very like Dropbox in this). You will also be able to access items saved to iCloud Drive on your Mac, iPhone, or iPad within this folder on your PC using File Explorer. Fully searchable, the service also lets you share your files with others.

How do I access iCloud using a browser?

At those times when it makes sense to use the browser to access your iCloud data — such as when travelling and using a borrowed machine.  It’s easy: just navigate to iCloud.com using your browser and log in using your Apple ID and password. 

In a neat touch, any active Reminders you might have will appear at the top of the iCloud main page when you log in. You’ll also find an icon-based set of links to take you to your  iCloud apps and services, including your iCloud Drive.

• Mail
• Contacts
• Calendar
• Photos
• iCloud Drive
• Notes
• Reminders
• Find iPhone
• Account Settings

You’ll also find fully functional online versions of Pages, Numbers and Keynote, which you can use to create documents compatible with both Mac and PC. You may also find News Publisher, if you have enabled that app.

NB: There are a small number of useful tasks you can only achieve using iCloud via your browser, such as creating rules to help improve email management across all your systems.

Accessing iCloud using a browser is especially useful on borrowed machines, during emergencies, while visiting Internet cafes or even using a machine you don’t own that you happen to have access to. Just remember to log out.

Of course, the beauty of iCloud is its ability to sync across all your devices, but to sync from a Windows system you need to install iCloud for Windows.

How to set up iCloud for Windows

Apple’s iCloud for Windows software should install automatically once downloaded. If it does not, open File Explorer, launch iCloud Setup and restart your PC. 

Setting up the app will create iCloud Photos and iCloud Drive folders on your Windows 10 device. These will sync with your iCloud account and connected devices, and should automatically appear in File Explorer’s Quick Access menu.

What are the system requirements?

The current iteration of iCloud for Windows requires that you are running at least Windows 10 and that you re signed into your Microsoft account. You’ll also need Outlook and a web browser (Microsoft Edge, Firefox, Chrome).

If you use Windows 7 or Windows 8, you can download compatible versions of iCloud for Windows directly from the Apple website. (Some iCloud services might work differently, as detailed here.)

Once the software is installed and open, you’ll be asked to enter your Apple ID to sign into iCloud, and you’ll then need to choose which iCloud features you want to use. To do this, you’ll be shown a window in which you check or uncheck each feature:

• iCloud Drive,
• Photos & videos, including sharing albums of images and videos.
• Mail, Contacts, Calendars, Tasks
• Bookmarks.

As you enable each of these categories, iCloud for Windows will create folders for those categories in File Explorer, so you can dive in to access or add to your iCloud content. You can also share and collaborate on files held in your iCloud Drive from within File Explorer, with edits synced across your devices.

Here is how to use each of these features:

How does iCloud Photos work?

When enabled, iCloud for Windows creates an iCloud Photos folder in File Explorer. Any images of videos you add to this folder will then be synced across all the devices you own that logged into this Apple ID. You can share galleries of images using iCloud Photo Sharing.

iCloud for Windows will keep lightweight copies of your images on your PC, only downloading full-res versions on request.

When you set yourself up with iCloud for Windows you’ll see an ‘Options’ item appear beside Photos. Tap this and you’ll find the following choices:

• iCloud Photo Library: Enable this and any images/videos saved to the iCloud Photos folder on your Windows device will be stored in iCloud. You can also change which folders are used for photo and video sync.
• My Photo Stream: Downloads all your most recent images from your devices to your Windows system.
• Download new photos and videos to my PC: This option appears when you enable iCloud Photo Library, and is self explanatory.
• Keep high-efficiency originals if available: Check this if you have plenty of drive space.
• Upload new photos and videos from my PC: Disable this if you only want to view items captured by your other iCloud devices.
• iCloud Photo Sharing: You can view items shared with you by others.

Mail, Contacts, Calendars, Tasks and Bookmarks

The final selection of iCloud items you can access using Windows. Once you enable this all your iCloud mail (typically sent to [email protected]) will be made available in the folder pane on Outlook. You’ll also find all your other calendaring data there, and bookmarks will be made available through your browser.

Apple recently made an extension available for the Chrome browser on Windows to ensure it remains compatible with iCloud and iCloud for Windows. Once the new version is installed, users should find a new “Passwords” section in the app with an iCloud Keychain logo.

Everything else

If you have any problems with these features, please refer to Apple’s extensive Support pages that should help you resolve them. (Don’t forget, you won’t receive email if you failed to create an iCloud email address when setting up your Apple ID).

There are some iCloud services that just don’t work on Windows: Find My iPhone, Back to My Mac, Backup, Notes and Reading List. All these require deeper OS integration, though in the case of Find My iPhone you will be able to track lost devices using iCloud in your browser.

However, if you have valuable data that you want to sync between your Macs, iOS devices and Windows PC, iCloud for Windows should help – though that 5GB of storage Apple supplies remains miserly at best.

(Find out more on how to synchronize iCloud — and what to do when it won’t sync.)

Can I use a Managed Apple ID?

Unfortunately, iCloud for Windows isn’t yet supported if you use a Managed Apple ID.

More useful iCloud tips

Here is a short collection of what may be useful tips for iCloud users on any platform, not just Windows:

• Even more useful iCloud storage tips
• How to fix iCloud when it stops working
• 5 clever little iCloud tips you’ll probably use
• 10 Mac tips for Windows users
• How and why to use Rules in iCloud Mail
• How to fix iCloud sync in seconds

Got a story? Please drop me a line via Twitter and let me know. I’d like it if you chose to follow me on Twitter so I can let you know about new articles I publish and reports I find.

Microsoft this week released 49 updates (including two recent additions) on Patch Tuesday with no reported zero-day flaws, public disclosures, or newly released working exploits for the Microsoft ecosystem. This came as welcome news and is paired with low-risk changes to Microsoft Office. The company’s development platforms saw minor updates to Visual Studio, and both SQL Server and Microsoft Exchange were patch free for the month.

The team at Readiness has provided a useful infographic outlining the risks associated with each of the updates. 

Known issues 

Each month, Microsoft publishes a list of known issues that are part of the latest update cycle, including the following reported minor issues:

• After you install KB5034203 (dated 01/23/2024) or later updates, some Windows devices that use the DHCP Option 235 to discover Microsoft Connected Cache (MCC) nodes in their network might be unable to use those nodes. Microsoft is still working on this one. In the meantime there is a workaround that involves setting the Cache Hostname to 1. 

We recognize and respect Microsoft’s recent efforts with artificial intelligence (note, I did not say “AI” as that is an Apple thing now) but it would be nice if Microsoft resolved the profile picture (that you can’t change) known issue soon. 

Major revisions 

Microsoft published the following major revisions to past security and feature updates including:

• CVE-2024-30080: (see below for mitigations). This patch was updated late in the June release cycle. As this was an information update, no further action is required, unless you want to action the Microsoft recommended mitigations.

Mitigations and workarounds

Microsoft published the following vulnerability-related mitigations:

• CVE-2024-30070: DHCP Server Service Denial of Service Vulnerability. Microsoft (helpfully) notes that if you’re not using DHCP, you are not affected by this potential vector for DDOS attacks. 
• CVE-2024-30080: Microsoft Message Queuing (MSMQ) Remote Code Execution Vulnerability. Message Queuing security issues are tough to find, mitigate and test, so this might need some careful attention from your internal developers. At the very least, ensure that you have changed your ports from the MSMQ listening default (1801) to help reduce your attack surface. Microsoft also recommends you check to see whether the MSMQ HTTP-Support feature is enabled.

The team at Readiness analyzed the latest Patch Tuesday updates to provide detailed, actionable testing guidance based on assessing a large application portfolio and a detailed analysis of the Microsoft patches and their potential impact.

For this cycle, we have grouped the critical updates and required testing efforts into different functional areas including:

Microsoft Office

• Microsoft SharePoint will require basic document opening and multi-user access tests this month.

Microsoft .NET and Developer Tools

• There are no updates to Microsoft .NET requiring application portfolio testing this month.

Windows

The following core Microsoft features have been updated:

• Changes to Secure Boot will require testing of all third-party drivers.
• Code integrity policies need to be verified for Windows Lockdown (WLDP), Windows Defender Application Guard (WDAG) and the Windows Driver Policy for Intune deployments. We recommend you test your Windows desktop sandbox and ensure that it boots correctly.
• Changes to Windows networking will require testing at least two DHCP servers.
• Remote desktop-related updates will require VPN connection tests. Try some administrative commands from the Microsoft Management console (MMC) such as adding, connecting and disconnecting VPN connections.

This month’s update also affects several core systems such as Kernel32 and Win32K.SYS sub-systems. Unfortunately, these changes affect how applications behave at a fundamental level, which makes testing not just hard, but broad and expansive across your application portfolio. The Readiness team suggests that the following general application tests be performed against all of your core line-of-business applications.

• Test as many windows and pop-ups as possible.
• Check window title bars for errors, or poorly formatted text.
• Check for unusual items in the Windows taskbar.
• Thoroughly test File explorer (sorry about that).
• Test multiple applications, with multiple windows.

Automated testing will help with these scenarios (especially a testing platform that offers a “delta” or comparison between builds). However, for your line-of-business apps, getting the application owner (doing UAT) to test and approve the results is essential. 

Windows lifecycle update 

This section contains important changes to servicing (and most security updates) to Windows desktop and server platforms.

• Windows 10 Enterprise and Education, Version 21H2 will no longer be serviced as of June 11, 2024

For those planning ahead, Oct. 8, 2024, is a big day as Microsoft will no longer offer general servicing for the following desktop platforms:

• Windows 11 Enterprise and Education, Version 21H2
• Windows 11 Home and Pro, Version 22H2
• Windows 11 IoT Enterprise, Version 21H2

Each month, we break down the update cycle into product families (as defined by Microsoft) with the following basic groupings: 

• Browsers (Microsoft IE and Edge)
• Microsoft Windows (both desktop and server) 
• Microsoft Office
• Microsoft Exchange Server 
• Microsoft Development platforms (ASP.NET Core, .NET Core and Chakra Core)
• Adobe (if you get this far) 

Browsers

Microsoft has released seven minor updates to the Chromium-based browser (Edge), while the Chromium project has added six additional updates this week. These updates should have minor to negligible impact on applications that integrate and operate on Chromium. Add these updates to your standard patch release schedule. 

Windows

This month, Microsoft released one critical update (CVE-2024-30080) and 32 patches rated as important for Windows, covering the following key components: 

• Windows Win32 Kernel Subsystem, GRFX and drivers
• Networking (Wii-fi) and DHCP
• Storage and Error Reporting
• Crypto and BitLocker

The critical-rated patch relates to the core, but not often used, Message Queuing service (MSMQ) that could affect internal applications. Unusually, this patch has already been updated since the main release on Tuesday. That said, the Readiness team believes all these Windows patches can be added to your standard release schedule.

Microsoft Office 

There were no critical updates for Office this month, and only five patches rated as important. All five have low potential for exploitability (no worms, add-in vulnerabilities or Word macro issues) and should be added to your regular Microsoft Office update schedule.

Microsoft Exchange Server 

No updates for Microsoft Exchange Server or SQL Server this month, which, of course, is a good thing. 

Microsoft development platforms 

Microsoft released just three updates to Microsoft Visual Studio. These patches affect versions of the Microsoft developer platform from 2017 to 2022. All of the proposed changes are low risk and application specific. Add these updates to your standard developer release schedule.

Adobe Reader (if you get this far) 

We are back to the usual state of things, and Microsoft has not chosen to include any Adobe products this release cycle. This is a very good thing.

While the arrangement between OpenAI and Apple is attracting a lot of attention, Apple has put together a sizable number of its own large language model (LLM) tools that will run on a compatible device or in its secure cloud, Private Cloud Compute. 

Apple Vice President Craig Federighi calls Apple Intelligence, “the personal intelligence system that puts powerful generative models right at the core of your iPhone, iPad, and Mac.”

To achieve this, it draws on what your device knows about you and on-device intelligence, or, where necessary, in the cloud via the highly secure Private Cloud Compute system. At all times, Apple says it’s working to protect user privacy, which means your data is protected unless you choose to use a third-party AI, such as ChatGPT. 

In making these solutions, Apple has paid particular heed to creating tools that offer truly useful help. The critical idea is that they get things done for you without getting in the way of the easy user interaction you usually enjoy with your Apple product. 

Tools to help you write better

To help you write, Apple Intelligence can proofread and rewrite your text anywhere across your system, including within third-party apps. Apple Intelligence will also summarize a meeting transcript, long email, and website content; pretty much any large block of text can be made bite-sized.

Be warned, for some of these functions Apple Intelligence might need to use ChatGPT, but you’ll be told if that is the case and can cancel the request rather than sharing your information with a third-party service provider.

Mail is getting better

We wrote a little more to explain how Mail works here.

• The system works to figure out which of your incoming emails are most essential and places those emails at the top of your Inbox.
• AI will also create what it thinks are appropriate replies for you — you don’t have to use them and do get to approve them before they are despatched.

Meetings, now with AI assistants

Tap record when making a call or when inside a Note to capture audio recordings and transcripts. Once the call or meeting ends, Apple Intelligence will quickly generate a summary of the transcript.

Tools to help you stay focused

There’s a new automated Focus mode that reduces interruptions but is also intelligent enough to let important notifications break through. Apple Intelligence will also get to know which of your notifications matter to you most and make sure those are at the top of your notifications list. The idea is to optimize your attention so you can stay on top of the things.

Making images

Apple’s on-device LLM engine will create original images for you based on a typed request. Usefully, it will also remove unwanted objects in an image on request. And a new Image Playground app lets you experiment with ideas and try different image styles to create your own images.

Photos gets better at helping you find your stuff

AI features in Photos include far more powerful and contextually-aware Search results and the ability to create a Memories video based on such a search.

Introducing, Genmoji

If like me you have problems finding precisely the right emoji or aren’t really certain if any that you do choose to have a double meaning, then salvation is at hand! Genmoji makes it possible to create completely original emoji on demand; just tell your Apple device what you want and up it will pop.

Wave your Image Wand

This feature needs an Apple Pencil. It works like this: Open a Note, draw a circle where you want your generated image to appear in that Note, and Apple’s intelligence will make you a custom image that reflects the contents of it.

Siri gets serious attention

We’re being promised lots of improvements in Siri; not only will it be able to better understand more complex or poorly articulated requests, but it also gains the kind of contextual understanding you need to figure out answers to complex questions such as “Show me the recipe Sacha sent me the other day.” 

That has several implications, including:

• Siri knows what you are looking at and you can make requests that reference that, such as adding an address to your Contacts, or adding something to a note in a different app.
• The assistant can also now answer questions about any of your Apple devices or operating system features, like an Apple Genius in your pocket.
• Siri also now understands typed requests — double tap the bottom of the display and a keyboard pops up.
• Sometimes your device might need to use ChatGPT to fulfill some requests; you will be told if that is the case and can cancel the request.
• Apple has also given asking Siri questions a new vibe; when you do so, your device now will show a glowing light all around the borders of the screen. 

Is there more to come? Probably

It is likely there will be additional features in place by the time Apple Intelligence is made available in the fall product software updates. This is because developers can use App Intents to make features available within their apps also available across the system. Meanwhile, developers get to use Xcode Complete to work smarter.

Apple is also thought to be pushing other genAI firms beyond OpenAI to offer their services on its platforms, while the company hopes to generate new income streams as developers build and make available additional LLMs on its platforms.

Please follow me on Mastodon, or join me in the AppleHolic’s bar & grill and Apple Discussions groups on MeWe.

When you think back to the heady days of last week, it’s hard to ignore that the introduction of Apple Intelligence on Monday now means the company formerly known as “behind on AI” just overtook Microsoft to bring home the world’s biggest and most trusted platform for generative AI (genAI).

The reason it’s hard to ignore is because it is true. Sure, Microsoft Copilot does lots of things already and once you can actually spend cash on Surface devices that run Copilot, Microsoft will be a powerful AI platform.  But Apple will be, too. (You can understand why Microsoft was unhappy about Apple’s deal with OpenAI.)

Mac, mobile, tablet, cloud

But I’d argue that Apple might be more powerful, because Apple Intelligence (the name Apple has given its own genAI models, served up both on device and in the cloud) will run on PCs (Macs), tablets, and mobile devices. 

I imagine it will in the future be supported on visionOS, and suspect it is only a viable usage case away from running on any other Apple device equipped with a capable processor. 

Millions of users on day one

Once Apple Intelligence is released, it will be available on hundreds of millions of devices on day one. Those devices won’t need to be new, either — those are the ones that are already being used. Apple says any Mac or iPad running an M1 or later chip will run it. It also says any iPhone 15 Pro series model will be able to handle it.

That’s a huge market, and no one needs to spend a dime to get into it. Those who do plan to upgrade or want to invest in exploring what’s available have plenty to look forward to on the journey to fall when the upgrade ships.

After all, if you squint, you can just about see M4 Macs appearing at the top of that nearby hill now the iPad Pro has that chip. We also think new iPhones are very likely to be already rolling off production lines, and both the iPad Air and iPad Pro already host compatible chips.  So, if you’re in the market for new hardware, it’s on the way, but tens of millions of people won’t need to upgrade to try Apple’s new genAI shiny.

Competitors are gasping

When it comes to Apple Silicon — the fire horse on which AI rides in the newly rebirthed Apple universe — take note that it has taken competitors, even those working with Arm reference designs, literally years to even begin to catch up with the computational power and energy efficiency Apple’s designers achieved. 

With a road map to M4 and beyond already in place, Apple is quite evidently telling us that when it comes to processors the speed argument is done, dusted, and won. And don’t get me started on privacy after the Microsoft Recall farce.

The power of consent

We can see Apple becoming the world’s leading provider of genAI-wielding devices for the mass market. Not only that, but because it is building a system that is private by design, consumers get the best of both worlds: complete data privacy for some tasks and the option of using AI from other vendors for other tasks, if they choose to do so.

That’s an alluring combination of privacy and consent, and that choice — that sense of personal AI-boosted empowerment — is coming to literally hundreds of millions of devices in real time. (Though that number is dented by the fact that we believe only US devices will get these features in the first draft, which is a shame.)

Consumers don’t have to use AI, but they will be happy that they can.

But it’s also about making 

Apple Intelligence can handle a huge number of tasks. Boosted by developers working with their own AI models and App Intents, the number of tasks it can do will only grow. 

I mentioned that developers will be able to use their own AI models in apps. What do you think they will build those models on? Well, they are Apple developers and will very likely build them using a Mac. 

That won’t just be because you need a Mac to build Apple apps, but also part because the M-series chips inside Apple’s computers demonstrate excellent training performance using most AI training tools. That’s why developers building AI models for other platforms also use Macs.

Now, I don’t know if the MacBook Pro is the most popular system on which to build AI models among data scientists — you have to be cautious about cause and effect — but I think it’s probably close.

Controlling the means of production

You see, Apple doesn’t just build fantastic machines to build AI models on, but now also offers a vast global ecosystem of devices on which to run those AI models.

Ever since the genAI bubble blew up, Apple has flexed its innovative power to jump from somewhere at the back to head of the pack when it comes to the technology. It is not the first time it has made a leap like that — it didn’t build the first music player, smart mobile device, netbook (it skipped them for tablets), or music streaming service, either.

But despite the impressive way in which Apple has jumped from zero to hero, the inconvenient question remains: Are humans ready for AI and the gigantic leaps in machine intelligence it will bring? Particularly when you look ahead to when we will run AI on quantum processors, as we eventually will — what have we unleashed?

Please follow me on Mastodon, or join me in the AppleHolic’s bar & grill and Apple Discussions groups on MeWe.

Adobe will change language in the terms of use for its software, as it seeks to clarify its position on content ownership and the use of customer data to train generative AI (genAI) models. 

The move comes in response to a customer backlash over an update to Adobe’s terms of use for Creative Cloud and Document Cloud software earlier this month. Some customers had interpreted the language to mean Adobe would gain access to their work for training purposes. 

The updated terms include several changes, including a statement that Adobe “may access your content through both automated and manual methods, such as for content review.”

Subscribers were required to re-accept the updated terms or lose access to Adobe’s apps and services. 

Customers raised concerns over Adobe’s ability to view and access content, highlighting concerns aboutcontent protected by a non-disclosure agreement.

Adobe attempted to explain the changes to its terms in a blog post June 6, stating the intention was to clarify updates to its moderation processes. Adobe said that it does not train its Firefly genAI model on customer content, and will “never assume ownerships of a customer’s work.”

On Monday, the company published another post, this time acknowledging the need to clarify the language in its terms of service after customer pushback. Adobe said it would consult with customers before changes are made on June 18. 

Adobe said that while its stance around the use of customer content is unchanged, it should have “modernized” and updated the terms of use sooner to explain legal language more clearly to customers. 

“Our updated Terms of Use, which we will be releasing next week, will be more precise, will be limited to only the activities we know we need to do now and in the immediate future, and uses more plain language and examples to help customers understand what they mean and why we have them,” Adobe said in the latest  blog post. 

While nine out of 10 IT professionals say they want to implement generative artificial intelligence (genAI) in their organization, more than half have integration, security and privacy concerns, according to a recent survey released Wednesday by Solarwinds, an infrastructure management software firm.

The SolarWinds 2024 IT Trends report, AI: Friend or Foe? found that very few IT pros are confident in their organization’s readiness to integrate genAI. The company surveyed about 7,000 IT professionals online regarding their views of the fast-evolving technology, and despite a near-unanimous desire to adopt genAI and other AI-based tools, less than half of respondents feel their infrastructure can work with the new technology.

Only 43% said they are confident that their company’s databases can meet the increased needs of AI, and even fewer (38%) trust the quality of data or training used in developing the technology. “Because of this, today’s IT teams see AI as an advisor (33%) and a sidekick (20%) rather than a solo decision-maker,” SolarWind said in its report.

Privacy and security worries were cited as the top barriers to genAI integration, and IT pros specifically called for increased government regulations to address security (72%) and privacy (64%) issues. When asked about challenges with AI, 41% said they’ve had negative experiences; of those, privacy concerns (48%) and security risks (43%) were most often cited.

More than half of respondents also believe government regulation should play a role in combating misinformation. “To ensure successful and secure AI adoption, IT pros recognize that organizations must develop thorough policies on ethics, data privacy, and compliance, pointing to ethical considerations and concerns about job displacement as other significant barriers to AI adoption,” the report said.

SolarWinds found that more than a third of organizations still lack ethics, privacy and compliance policies in place to guide proper genAI implementation. “While talk of AI has dominated the industry, IT leaders and teams recognize the outsize risks of the still-developing technology, heightened by the rush to build AI quickly rather than smartly,” said Krishna Sai, senior vice president, technology and engineering, at SolarWinds.

Indeed, leading security experts are predicting hackers will increasingly target genAI systems and attempt to poison them by corrupting data or the models themselves. Earlier this year, the US National Institute of Standards and Technology (NIST) published a paper warning that “poisoning attacks are very powerful and can cause either an availability violation or an integrity violation.

“In particular, availability poisoning attacks cause indiscriminate degradation of the machine learning model on all samples, while targeted and backdoor poisoning attacks are stealthier and induce integrity violations on a small set of target samples,” NIST said.

Overall, the IT industry’s sentiment reflects “cautious optimism about AI despite the obstacles,” SolarWinds reported. Almost half of IT professionals (46%) want their company to move faster in implementing the technology, despite costs, challenges, and concerns, but only 43% are confident that their company’s databases can meet the increased needs of AI. Moreover, even fewer (38%) trust the quality of data or training used in developing AI technologies.

IT pros cited AIOps (Artificial Intelligence for IT Operations) as the technology that will have the most significant positive impact on their role (31%), ranking above large language models and machine learning. More than a third of respondents (38%) said their companies already use AI to make IT operations more efficient and effective.