Adobe’s AI model for video generation is now available in a limited beta, enabling users to create short video clips from text and image prompts.

The Firefly Video model, first unveiled in April, is the latest generative AI model Adobe has developed for its Creative Cloud products — the others cover image, design and vector graphic generation.

From Monday, there are two ways to access the Firefly Video model as part of the beta trial.

One is the text and image to video generation that Adobe previewed last month, accessible in the Firefly web app at firefly.adobe.com. This enables users to create five-second, 720p-resolution videos from natural-language text prompts. These can contain realistic video footage and 2D or 3D animations. It’s also possible to generate video using still images as a prompt, meaning a photograph or illustration could be used to create b-roll footage.

To provide greater control over the output, there are options for different camera angles, shot size, motion and zoom, for example, while Adobe says it’s working on more ways to direct the AI-generated video.

Waiting list

Adobe said it only trains the video model on stock footage and public domain data that it has rights to use for training its AI models. It won’t use customer data or data scraped from the internet, it said.

To access the beta, you’ll need to join the waitlist. It’s free for now, though Adobe said in a new release that it will reveal pricing information once the Firefly Video model gets a full launch.

Adobe is one of several technology companies working on AI video generation capabilities. OpenAI’s Sora promises to let users create minute-long video clips, while Meta recently announced its Movie Gen video model and Google unveiled Veo back in May. However, none of these tools are publicly available at this stage.

Extended remix

The other way to access the Firefly Video model is with the Generative Extend tool, available in beta in video editing app Premiere Pro. Generate Extend can be used to create new frames to lengthen a video clip — although only by a couple of seconds, enabling an editor to hold a shot longer to create smoother transitions. Footage created with Generative Extend must be 1920×1080 or 1280×720 during the beta, though Adobe said its working on support for higher resolutions.

Background audio can also be extended for up to 10 seconds, thanks to Adobe’s AI audio generation technology, though spoken dialogue can’t be generated.

At its MAX conference on Monday, Adobe also announced that its  GenStudio for Marketing Performance app, designed to help businesses manage the influx of AI-generated content, is now generally available.

Adobe’s GenStudio content supply chain platform is now generally available, with the ability to publish content directly to social media channels such as Instagram, Snap and TikTok coming soon.

Adobe launched GenStudio for Performance Marketing — as the standalone GenStudio application is now called — in preview at Adobe MAX conference in 2023. At this year’s MAX event it made a slight change in branding: GenStudio now refers to both the GenStudio for Performance Marketing app and the various Adobe applications it integrates with, such as Adobe Experience Manager, Adobe Express, and Workfront.

Adobe has been quick to integrate its Firefly generative AI models across Creative Cloud apps such as Photoshop and Illustrator, enabling designers to increase their output significantly, the company says. (IDC analysts also predict genAI will boost marketing team productivity by 40% in the next five years.)

The aim of GenStudio for Performance Marketing is to help marketers access and use the AI-generated content created within their organization while respecting brand guidelines and legal compliance policies.

“The challenge facing most brands out there is that they have an inefficient content supply chain, where bottlenecks appear in areas like planning, content development and measurement,” said Varun Parmar, general manager for GenStudio at Adobe, in a news briefing. This is where GenStudio for Performance Marketing can help, he said, providing a “seamless way for brands and agencies to deliver on-brand and personalized content that is compliant with brand standards.”

GenStudio for Performance Marketing performs several functions. First, it serves as a content repository where users can access pre-approved assets such as images, logos, and videos for use in the creation of marketing content. This could be anything from display ads to banners and emails. To enable reuse of content across campaigns, GenStudio for Performance Marketing integrates with Adobe Experience Manager Assets, Adobe’s digital asset management app.

Firefly video

Users can also edit and adapt existing assets from the app using the Firefly AI models. This could mean creating variations of email ads tailored to a specific geographic region, for instance.

Those models will soon include new video capabilities, including text-to-video and image-to-video, now available as beta versions.

In GenStudio for Performance Marketing, an AI-powered “brand check” feature can automatically inspect assets before they are used in marketing campaigns, comparing with pre-defined templates and alerting marketing and design teams where content may be out of step with a firm’s brand compliance guidelines. Here, each asset is given a score out of 100, with detailed recommendations for changes: an email headline that’s too lengthy, for example, or innappropriate tone of voice. An integration with Adobe’s Workfront also enables automated “multi-step review workflows,” to provide additional oversight of the approval process

Adobe also plans to let users publish content directly from GenStudio for Performance Marketing to social media channels from the likes of Meta, TikTok and Snap, as well as display ad campaigns with Google’s Campaign Manager 360, Amazon Ads and Microsoft Advertising. This campaign activation feature is “coming soon,” Adobe said, without providing further details. It will also be possible for customers to publish content via their own email and web channels via Adobe Journey Optimizer in future, Adobe said.

Finally, GenStudio for Performance Marketing will provide analytics on the performance of content that’s live on platforms owned by Meta (such as click-through rate, cost per click and spend), with integrations with others such as Microsoft Advertising, Snap and TikTok also available “soon.”

“All companies have to ramp up their genAI knowledge and its impact on brand content/assets,” said Jessica Liu, principal analyst at Forrester.

“Solutions like GenStudio present compelling opportunities for companies to alter their creative development and production process — such as creating more content, accelerating workflows, streamlining workflows, or shifting workforce skillsets.”

Ad customization comes at a customized price

Adobe hasn’t published a list price for GenStudio and GenStudio for Performance Marketing. A company representative said, “As this is enterprise software, there isn’t a one size fits all pricing as it’s based on the customer need/requirement.”

AMD’s big Advancing AI event in San Francisco on Thursday underlined how quickly the microprocessor industry has pivoted to artificial intelligence (AI) as its main sales pitch. 

The company offered three hardware announcements across its processor line-up, each appealing to different parts of the AI market.

The first was the new Instinct MI325X AI accelerator chip, a datacenter-oriented GPU which ups performance on every metric compared to last year’s MI300. The company also showed off its fifth-generation EPYC processors for the enterprise cloud and datacenter sector. And it unveiled the new Ryzen AI PRO 300 series, a family of processors for mobile PCs aimed at enterprise buyers.

Mobile chips have traditionally been low power (and lower performance versions) of their desktop equivalents, but with the focus on AI, that distinction is fast disappearing.

AI requires more raw power, which is now showing up in the specifications for new chips. For example, the Ryzen AI PRO 300 series offers three processors, starting with the Ryzen AI 9 HX PRO 375; it features 12 Zen 5 cores/24 threads, a clock speed that can be boosted to 5.1GHz, and integrated Radeon 890M graphics. 

It also features a neural processing unit (NPU) that delivers up to 55 tera operations per second (TOPS), making it the most powerful desktop AI chip of its kind on the market. 

Just below that in the line-up is the Ryzen AI 9 HX PRO 370, an identical chip with slightly less NPU performance — up to 50 TOPS. And the entry-level chip is the Ryzen AI 7 PRO 360 with 8 cores/16 threads, a 5GHz clock speed, Radeon 880M graphics, and the same 50 TOPS NPU performance.

NPUs are on the new frontline of desktop competition because they make possible features such as accelerating Microsoft’s Copilot + and AI-intensive tasks such as real-time language translation.

They also pit AMD against traditional rival Intel, which has had NPUs of its own in its Core Ultra CPUs since 2023. Today, NPU AI acceleration is premium priced, but there are signs the technology is likely to jump quickly to more mainstream chips.

“With Ryzen AI, we’ve actually enabled hundreds of different AI functions,” said AMD CEO Lisa Su near the end of a two-hour Advancing AI presentation. “Our latest software stack makes it really easy for developers to optimize thousands of pre-trained models for Ryzen. 

“Our Ryzen AI Pro 300 series resets the bar for what a business PC can do,” she said.

AI-enabled chips could quickly become standard issue across all sectors. Earlier this year, IDC predicted that by 2027, PCs with AI acceleration built in would constitute 60% of all PC shipments.

On the same day as AMD’s event, Intel announced its new Core Ultra 200S processors, which the company claims deliver NPU performance of 36 TOPs between its CPU, GPU and NPU. A third player in the market offering NPU technology is Qualcomm, which launched its Snapdragon X PC chip in September. 

One of two plants that mine high-purity quartz and silica sand needed for the production of semiconductors and other high-tech hardware has reopened operations after being shut down for more than a week.

Sibelco today announced the restart of production at its Spruce Pine, NC mining and processing operations following the disruption caused by Hurricane Helene. Sibelco had previously announced that all its employees are safe.

“While the road to full recovery for our communities will be long, restarting our operations and resuming shipments to customers are important contributors to rebuilding the local economy,” Sibelco CEO Hilmar Rode said in a statement.

The devastation to the mountain town in western North Carolina by Hurricane Helene brought to a halt the mining and refining of high-purity quartz by both the Sibelco and The Quartz Corp mines in Spruce Pine, NC, the two primary sources of the precious crystalline mineral.

The Quartz Corp did not immediately have an update on its operation’s status today.

Sibelco, the far larger of the two mining companies in the region, said in a statement last week that its operations had no power, but its infrastructure had only sustained minor damage. “The repair of power lines leading to our plants has progressed significantly,” Sibelco said. “Our final product stock has not been impacted. We are working closely with our customers to assess their needs and plan the restart of product shipments as soon as we can.”

With ever increasing demand for more computer chips to power digital systems and services, last year Sibelco announced it would spend $200 million double its output of high-purity quartz.

The computer chip industry is already struggling to increase production of semiconductors in the US due to a impending lack of talent needed to run new fabrication plants now being built in the US.

Ultra-pure quartz, which formed millions of years ago in the Appalachian Mountain region, is used to create “crucibles” in which a pure polysilicon is melted down to be used in creating silicon chips. The ultra-high quality quartz can both withstand the extreme temperatures needed to melt silicon and ensures no impurities are introduced in the process.

From 70% to 90% of the crucibles used in silicon production worldwide are made from Spruce Pine quartz, according to a report.

The high‑purity silicon dioxide particles that result from processing by the plants are the raw materials from which computer chips and other high-tech hardware is made, including fiber‑optic cables and solar photovoltaic cells.

On Oct. 2, The Quartz Corp. said in a statement that most of the damage to its quartz processing plant was to “ancillary units.” May Kristin Haugen, a spokesperson for The Quartz Corp, told Computerworld it has “brought in numerous experts to assess the three different plants.

“All our three plants in Spruce Pine are affected, though in different ways,” Haugen said. “They are situated in different locations and the consequences for production will likely vary.”

Sibelco

Based on outside “expert assessment,” the company plans to communicate the status and restoration plans when it’s ready. “Our restoration plan will, however, depend on the surrounding infrastructure such as power, water, roads, and railway. Depending on the damage, it can take time to restore production, but we will get there,” she said.

The Quartz Corp said the COVID-19 pandemic taught it the value of “sizable safety stocks,” and between that and its multiple locations, it’s not concerned about shortages “in the short or medium term.”

Apple-focused IT admins using device management software should perhaps temporarily disable iPhone Mirroring across their device fleets to prevent inadvertent privacy or compliance challenges as a result of using the new macOS Sequoia feature. (Apple is supposed to be working on a fix already.)

It appears at present that when you use iPhone Mirroring with a Mac running Sequoia your computer gathers a small amount of information about the iPhone apps being used. It doesn’t gather all the data, just very basic information concerning app name, time of use, and so on — and while some MDM systems reportedly don’t parse this information, some of the most commonly used data compliance tools do review it.

What is the real threat here?

Ultimately the problem is two-fold: 

• Privacy: First, managed workplace Macs are gathering data concerning apps used on personally-owned iPhones, which can be a privacy failure and could be a bigger problem in some contexts. (For instance, an employee in an authoritarian state in which use of VPN or LGBTQ apps is proscribed might find their app use shared by this bug, with potentially serious consequences.)
• Compliance: The second problem concerns regulatory compliance: If a compliance audit tool picks up use of an unauthorized iPhone app on a corporate network, which they will do due to the architecture of this bug, IT will be forced to explain and look into that use. This poses enterprise-wide compliance challenges, and also means admins could be forced to waste time on what should be a relatively trivial problem.

The iPhone Mirroring SNAFU isn’t a problem for smaller firms that don’t use device management or compliance tools, as in theory at least, the information gathered is not made available to anyone but the registered Apple ID/user of a system. Though the fact the data exists at all might pose an additional attack surface for data exfiltration. 

What is the problem?

The snag was first spotted in late September by Sevco Security, a company that does not develop for the Mac. It found that when iPhone Mirroring is used, any iPhone app creates an entry in a library item on your Mac. Effectively that is because the Mac treats these apps as native Mac apps, even though they are being run on iPhone.

You can read an in-depth account of the behavior courtesy of Sevco (above), but essentially if you run the mdfind CLI (Command Line Interface) in Spotlight you should see a complete list of both iPhone and Mac apps run on the Mac. You usually can only see the Mac apps used, but with iPhone Mirroring you now see iPhone apps, too. That information is then maintained in a deeply-stashed library file on the Mac, which most users will never see.

The problem is that most compliance, network, and endpoint security and audit tools will interrogate the library files to discover what apps are being run, including apps run on the Mac via iPhone. (They can’t see any of thew app data but could still provide insights that threaten privacy or compliance.)

Apple is working on a fix

Apple is working on a patch for the flaw, but it doesn’t seem to have appeared in the latest beta. At the same time, it’s worth noting that rather than giving Apple 30 days to rectify the problem (which is the usual approach for revelations of this kind), Sevco disclosed the problem just 12 days after informing Apple of it, citing the public interest as many Mac users work with iPhone Mirroring.

Sevco did say: “We appreciate Apple’s rapid response and urgency addressing the issue.”

What you should do now

Sevco offers the following advice pending a fix:

• “Employees should not use iPhone Mirroring on work computers;
• “Companies should communicate to employees that they should avoid using iPhone Mirroring on work computers (this may be a legal or regulatory requirement);
• “Companies should identify any enterprise IT systems that collect software inventory from Macs and work with those vendors to mitigate the risk until a patch is available.”

It is important to stress that since Apple is working on a fix, this is unlikely to be a permanent concern. And most enterprises handling confidential data should already have forbidden the use of iPhone Mirroring on managed devices to prevent other forms of data exfiltration.

Switch it off and on again?

Some admins have noted that in cases in which such information has already been collected, getting users to log out of their Apple Account and login again might destroy the information held on the Mac. They can then disable iPhone Mirroring pending Apple’s fix. While logging out of an Apple Account seems a rather large hammer for a relatively small problem, if you are handling sensitive information, or have apps you don’t want to share the names of, it may be a useful step. (I’ve not tested this myself so cannot be certain this will completely wipe away the information.)

Should you panic?

This is not a red alert. Apple will rectify this problem soon, and its existence is unlikely to tarnish Apple’s reputation for security — certainly not in comparison to the appalling multi-billion dollars damage wrought by the recent Microsoft/Crowdstrike failure. While the flaw does pose compliance and privacy challenges, and the collection of the information itself flies in the face of Apple’s general promise to collect as little data as possible about what users do, it can be rectified.

At the same time, it is likely that Windows-invested security experts will redouble their attempts to poke holes in Apple’s reputation for security as they recognize the growing threat Apple now provides to the ecosystem in which they have so much invested. That’s particularly true now that Delta Airlines has hired David Boies’ feared law firm to pursue damages generated by the Crowdstrike mess. 

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

Adobe has unveiled plans for a new web app called Content Authenticity, which is part of the Content Credentials platform and will be used to ensure that AI models are not trained on content creators’ images, video, or audio clips without their permission.

The app can be used with all images and clips, not just those created with Adobe’s software, according to The Verge.

A public beta version of Content Authenticity will be released in the first quarter of 2025, the company said. All that is required is a free Adobe account, meaning users don’t need a paid subscription to use the web app.

Multiple research firms reported that the unemployment rate for IT workers dropped precipitously last month — in one instance, by nearly 37% — a decline that marked a serious shift from months of incremental increases mixed with plateaus in unemployment.

Tech industry watchers say the dramatic uptick in hiring is the result of several factors, including a strengthening economy, AI-related talent needs, and increased hiring by small-to-midsize companies.

Smaller organizations have also been scooping up talent left in the wake of more than two-years’ worth of layoffs by mega corporations. Last month, the number of unemployed IT professionals in the US dropped from 148,000 to 98,000, according to IT industry consultancy Janco Associates. Janco derived its findings from a US Bureau of Labor Statistics (BLS) report released last week.

More than 78,000 IT pros were hired in September, cutting into the backlog of unemployed. “IT Pros who were unemployed last month found jobs more quickly than was anticipated, as CIOs rushed to fill open positions,” said Janco CEO Victor Janulaitis. “Our analysis predicts the same will be the case for the next several months.”

Janco pegged the September unemployment rate for IT workers at 3.8%, down from 6% in August. IT unemployment is now below the overall national unemployment rate of 4.1%, according to Janco.

Over the year, Janco has reported IT unemployment rates that were far higher than other organizations’ figures. For example, nonprofit IT association CompTIA said tech unemployment dropped from 3.4% in August to 2.5% in September; that still represents a 26.5% drop.

Janco’s numbers are calculated differently from CompTIA’s and the firm’s report showed the overall unemployment rate for IT pros grew from 5.6% in July to 6% in August. (High unemployment is defined by the BLS as being 5.5% or greater.)

Janco’s figures, in fact, showed IT unemployment had surpassed the national unemployment rates for seven of the last eight months. But in September, that shifted dramatically.

“There are now approximately 4.18 million jobs for IT professionals in the US,” said Janulaitis. “Layoffs at big tech companies continued to hurt overall IT hiring. Large high-tech firms continue to lay off to have better bottom lines. Included in that group of companies that have recently announced new layoffs are Intel, Microsoft, and Google.”

Janco’s current forecast is for the IT job market to grow slightly, between 5,000 and 6,000 jobs the remainder of 2024. That, however, will not be enough to absorb all unemployed IT pros looking for work.

AI part of the hiring boom?

Martha Heller, CEO of Heller Search, a tech executive headhunter firm, said a significant portion of the $600 billion that has already been invested in AI technologies by organizations is going to talent — and not just AI talent. “To get ROI from AI, most companies need to hire more data engineers, cybersecurity leaders, and developers in addition to modelers and prompt engineers,” Heller said.

Through 2027, generative AI (genAI) will spawn new roles in software engineering and operations, requiring 80% of the engineering workforce to upskill, according to research firm Gartner. A Gartner survey of software engineering leaders found that 56% rated AI and machine learning engineer as the most in-demand role for 2024.

While AI will make engineering more efficient, organizations will need even more skilled software engineers to meet the rapidly increasing demand for AI-empowered software. “Bold claims on the ability of AI have led to speculation that AI could reduce demand for human engineers or even supplant them entirely,” said Philip Walsh, a Gartner senior principal analyst. “While AI will transform the future role of software engineers, human expertise and creativity will always be essential to delivering complex, innovative software.” 

IT workers are paying attention to the need to bolster their skills. A survey by genAI tech company Amdocs of 500 full-time workers who use Gen AI-powered tools or software showed half of Gen Z respondents would consider looking for a new job if their company did not train them on the technology. That compared to 37% of Gen X and 27% of Baby Boomers who said the same thing.

“Many of the larger firms continue to be focused on improvements in productivity and replacing lower-level skills with AI applications,” Janulaitis said, noting that AI continues to slow the growth of entry-level IT positions, especially in customer service, internal reporting, telecommunications, and hosting automation.

C-level executives continue to focus on eliminating “non-essential” managers, staff, and services, he said. “Experienced coders and developers have limited opportunities with legacy applications. The highest demand continues to be for AI, security professionals, new technology programmers, and Internet processing IT pros.”

A closer look at BLS data for tech hires showed companies are pulling back on hiring AI pros and are instead seeking data researchers who can help businesses make better decisions — whether to advance AI deployments or broader business strategy.

Job openings for tech support specialists and database administrators were up 14%, the largest percent change for the month.

After nine consecutive months of growth, the total number of job postings for AI and machine learning engineers declined by 3.7% in September. And new job postings declined by 13.7% during the same period, according to Ger Doyle, head of Experis North America, a ManpowerGroup tech recruiting subsidiary. 

“This is mainly due to shifting demands. While there is less demand for software developers, there is increasing demand for roles such as solutions architects and data scientists to build robust data foundations,” Doyle said.