Feed aggregator

Creative work relies on technology. Instead of physical materials, a contemporary artist’s toolkit is largely digital, especially in the production world, where new resources emerge every day.

The repeating theme is AI, which creative studios are steadily adopting. It enables time-pressed artists to deliver work at speed, without sacrificing the human touch.

With the pace accelerating, keeping up with emerging platforms can give teams a real edge. Here are 12 AI tools redefining the stages of creativity – not as gimmicks, but as tangible working solutions.

1. Grip

(Image credit: Grip)

Speed and clarity are everything. Ideas need to quickly turn into pitch decks and deliverables otherwise, you risk losing client interest.

Grip enables artists to take a product and create its digital twin, ready for placement in any setting. Its ethically sourced generative AI model facilitates visual fine-tuning, allowing for the rapid creation of high-quality concept pieces.

2. Upscayl

(Image credit: Upscayl)

If you find yourself with pixelated reference images, open-source platform Upscayl converts them into a usable asset.

By restoring detail to low-res assets, teams can keep visual consistency without re-sourcing or recreating elements.

3. Viggle

(Image credit: Viggle)

A quick and easy way to share animations with clients is through Viggle, an AI motion capture platform that takes your footage and converts it into a stylised animation. Using generative AI prompts, your movement data can be transferred to any scene.

4. Leica Cyclone 3DR

With production underway, asset generation becomes the priority. Post-scan workflows can slow projects down, but Leica’s Cyclone 3DR addresses this by making it easier to segment, classify, and export individual assets from LiDAR captures.

It enables efficient environment building, especially when digital twins of real spaces are needed fast.

5. Vizcom

(Image credit: Vizcom)

Vizcom streamlines concept art for internal approvals or client reviews by converting sketches to 3D renders.

Using AI to retexture, recolour, and retouch, the software creates ready-to-use assets in minutes.

6. Gaussian Head Avatar

This research-driven tool creates 3D heads from just a photograph. Once created, users can customise hairstyle and proportions.

While not yet standardised in production, Gaussian Head Avatar opens up interesting applications for digital doubles, background characters, or stylised avatars that require volume but not photoreal precision.

7. Motorica

(Image credit: Motorica)

Once the assets are together, it’s time to begin animating. Browser-based Motorica requires 10 minutes of motion capture data to replicate an individual’s movement, then uses AI-powered prompts to generate consistent animation.

This makes it easier to prototype or expand performances without repeat capture, and is especially useful in stylised projects or large-scale simulations.

8. Colossyan

(Image credit: Colossyan)

When people-focused content is needed, but you don’t have a spokesperson, Colossyan presents a simple, AI-driven solution.

The platform generates an AI Avatar, and through a text-to-speech function, enables the avatar to lip-sync. It’s commonly seen in short-form content to add a personal touch – something brands are increasingly calling out for.

9. VASA-1

Sometimes, the best tools aren’t found in typical places. VASA-1 is a Microsoft research project that maps audio to facial animation.

It works for dubbing, creating content in different languages, or when you want to put visuals over a great soundbite. Although it’s not yet production-ready, it’s a tool to look out for.

10. Blend Now

(Image credit: Blend Now)

The main aim of AI tools is to cut down time for creatives, enabling their work to go further.

Blend Now does just that by removing backgrounds and replacing them with AI-generated ones.

While in-person shoots offer more depth and nuance, this is a simple way to repurpose content and keep it current for customers.

11. Beeble

(Image credit: Beeble)

Beeble does a similar task, but applies new backgrounds to videos. The AI-powered tool generates passes, removing the surrounding environment and all lighting.

The user applies new light sources according to the refreshed background, making the shot look natural and highly realistic – another option for those looking to refresh older footage.

12. Dream Machine

Edit elements within video – including lighting, wardrobe, and setting – using an AI inference.

The results tend to be mixed depending on the scale of the request, but the ability to test alternate creative directions without reshooting makes Dream Machine a useful option for short-form or commercial projects with rapid turnaround.

Navigate AI with intent

Above all else, AI tools are here to assist, not to replace creative work. Thanks to these new technologies, we can produce previews instantly, generate convincing videos for short-form content, and wrap up the editing process faster.

And as agentic AI draws closer, we’ll see even better ways to make creative work pop.

For now, the real advantage lies in careful selection. Forward-thinking teams are already testing and integrating these tools, using AI to reduce repetitive tasks and make space for important decisions.

And in an increasingly intense creative production landscape, modern technology is essential.

This article was produced as part of TechRadarPro's Expert Insights channel where we feature the best and brightest minds in the technology industry today. The views expressed here are those of the author and are not necessarily those of TechRadarPro or Future plc.

Supporters say the pop-up messages could encourage Minnesotans, especially kids, to think twice about how much time they spend on sites. Social media companies argue that the law is heavy-handed.

from the U.S. Education Department.
'/>

A program at the University of Oklahoma trains much-needed mental health professionals for rural schools in the state. Now, its federal grant funding is on the chopping block.

(Image credit: Jovana Mugosa for NPR)

Kids in the U.S. get most of their calories from ultra-processed foods, which are tied to health problems. Now, scientists are finding that kids don't all react to these foods in the same way.

(Image credit: Catherine Falls Commercial)

Most states will have to establish work requirements for Medicaid by 2027. Georgia has had them for two years. Some Georgians say the glitchy system makes it too hard to prove they actually qualify.

(Image credit: Jeff Amy)

Steve Inskeep speaks with former US Secretary for Transportation Pete Buttigieg about distrust in government and the status of the Democratic party.

(Image credit: Keren Carrión)

Once the exclusive domain of legal and compliance, data privacy is now a major responsibility for IT. When privacy issues emerge, IT professionals are on the spot, not necessarily as the team that absorbs privacy risk but the one accountable for the tools and visibility to proactively manage it.

When the board asks questions or when the auditors arrive, IT management must be ready with answers about checking the box for data privacy compliance, taking steps to mitigate third-party vendor risk, complying with reporting requirements and setting appropriate policies for data governance, especially around AI systems.

This is a tall order for data privacy and security and it’s compounded by the operational reality that available resources rarely match these expanding responsibilities.

Responsibilities and Reality

Several factors contribute to the disconnect between privacy expectations and operational capacities.

Privacy functions typically operate with minimal staffing while being organizationally siloed from IT operations. Yet IT leaders are expected to provide centralized accountability for privacy, yet they don’t always have the corresponding authority or visibility needed to support that responsibility.

Visibility. Data flows through IT systems, but its exact location and ownership often remain unclear. In other words, there’s no single source of data truth. Without unified monitoring of these complex pathways, privacy breaches become discoverable only after damage occurs.

Third Party Risk. Many privacy teams lack visibility into which vendors access personal data. This visibility gap creates compliance exposure that ultimately reflects on IT leadership and creates ownership confusion. Without protection and collaboration and shared visibility, things fall through the cracks.

Showing Proof. No one wants to get slammed with a violation for poor data handling. Authorities expect evidence that companies have taken steps to implement reasonable preventative measures –documentation many IT departments struggle to produce.

Dynamic data. Modern data environments create perpetual compliance challenges as data continuously streams in and throughout organizations. Maintaining accurate data inventories becomes nearly impossible through manual processes. Simultaneously, data subject requests (DSRs) are on the increase while records of processing activities (RoPAs) consume enormous legal and IT resources.

Furthermore, data integrations are a major concern. After all, who hasn't learned the hard way that not every catalog or pre-built integration supports every system, workflow, or business logic?

Best Practices

- Data mapping creates a dynamic data inventory that is direct and actionable for all involved to find clarity and purpose with enterprise data. Such an investor can facilitate the building joint workflows, clear responsibilities for data ownership so everyone is working through the same data inventory.

- Strive for more sophisticated automated behavior between tools and purpose. Automation helps to avoid repetitive tasks, delivers oversight, flags risks, track third party behavior and manages data integrations.

- In the aftermath of a breach, automation produces a record of that proves the checklist was complete, and appropriate steps were taken to avoid an incident. Auditors are less likely to issue steep fines for violations if IT can produce a report with this proof. Automation also frees up more time to analyze risk and refine processes.

- Take a no-code approach to integrations. No-code not only expands the number but enhances the quality of integrations. No-code enables each integration to be customized per the organization’s exact needs and on IT terms.

- Continue to focus on real-time visibility to enable the holy grail of the IT enterprise: monitoring and control

Helpful Capabilities

Data Mapping. Given the centralized role that CIOs and CISOs have for enterprise privacy, there’s a need strong collaboration and shared visibility with teams that are responsible for privacy yet have different priorities and reporting structures.

Data Mapping for inventory discovery and data classification are essential and typically done through a portal that provides a window on how data moves through the enterprises system and where personal data is accessed by which vendors.

By mapping and classifying data, a portal can enable one source of data truth for all aspects of privacy data and ensure that all privacy and legal teams are working with the same dynamic data inventory.

Automated Integrations. As more people exercise their data privacy rights, and more privacy mandates pass, the number of data subject requests has increased.

These requests are driving demand for automated processes that can keep pace with the time-consuming burden of foundational tasks like building and maintaining Records of Processing Activities (RoPAs) and Data Subject Requests (DSR). DSRs and ROPAs can result in data bottlenecks and constrained resources.

No Code Approach. Not only are privacy regulations complex and always changing, but the average organization also now connects to dozens of data sources, requiring big libraries of pre-built integrations and an API catalog rat race to build APIs.

When data reporting on this level starts to pile up, automated data integration becomes a game changer, so build data integrations that are easily facilitated with any backend system, platform, and SaaS apps. No-code integrations are a game changer in this respect. No-code allows IT teams to freely build, customize and maintain integrations that match internal systems, workflows, and logic - enabling faster deployment and easier maintenance of DSR handling, without developer overhead.

AI Agent. With all the complexity and intricacies of privacy management, there’s been an imbalance of resources and expectations, and it’s been ongoing for years. Automation has been helping to solve this imbalance, by cutting through the complexity. Now, core privacy tasks can be supported by an AI Assistant that can be purpose-built to automate core tasks, not just make suggestions.

AI agents are embedded with the privacy operations platform to intelligently analyze actual systems, how data is used and classified. It can help to build RoPAs automatically, freeing up valuable time for strategic initiatives. Beyond simply automating tasks, a privacy AI Agent can identify potential data risks, including shadow IT systems that lack necessary security controls to misclassified information.

Clear context on why these are risks along with actionable insights to help IT teams make informed decisions and mitigate potential issues proactively.

Summary

For IT, blind spots are not only a technical challenge but also organizational. Each exposure can be a chance to demonstrate strategic leadership by building greater trust with your team, users, privacy teams, and the board. Visibility also leads the way forward to getting ahead of regulatory changes.

Treating privacy blind spots seriously helps to build an agile, secure IT organization that is accountable, collaborative, and ready for growth. Forward-thinking IT leaders can turn compliance challenges into an operational advantage.

We've listed the best RPA software.

This article was produced as part of TechRadarPro's Expert Insights channel where we feature the best and brightest minds in the technology industry today. The views expressed here are those of the author and are not necessarily those of TechRadarPro or Future plc. If you are interested in contributing find out more here: https://www.techradar.com/news/submit-your-story-to-techradar-pro

Ceasefire talks have started between Thai and Cambodian leaders in Malaysia in an urgent effort to resolve deadly border clashes that entered a fifth day despite mounting international calls for peace.

(Image credit: The Government Spokesman Office)

AI agents, the much-touted next phase of generative AI, have commanded enterprises’ attention. Right now, 61% of business leaders are actively adopting AI agents, according to a recent survey by my organization – with ambitious plans to scale them organization-wide.

The fixation is justified: agents can work autonomously, navigate complex workflows, learn from experience, and leverage other software as tools. They are a step change from AI that talks with you, like chatbots, to AI that works for you. The result is major productivity: Gartner estimates that by 2028, agents will automate 15% of day-to-day business decisions.

But business leaders shouldn’t mistake agents’ sophistication with omnipotence. Agents can fall into the same traps as older, less sophisticated software – including dreaded IT silos.

Battling silos

For decades, IT professionals have battled silos: applications, databases, and other systems that aren't interoperable. In the 1980s and 1990s, enterprises struggled to connect disparate applications into a single ERP solution.

Accounting, procurement, and sales workstreams were stubbornly separate – squandering coveted cross-company insights. More recently, enterprises have struggled to unify crucial customer data across disparate CRMs, and have also labored to integrate data spread across on-premise locations and multiple cloud environments.

No matter the decade or technology, the result of silos is always the same: wasted time, wasted resources, and wasted potential. When agents become trapped in silos, the outcome is no different. Their return on investment plummets, too.

We are already seeing agentic silos take shape. Enterprises are using agents with rigid divisions – one agent for sales activities, another for procurement tasks, a third for CRMs – with little connective tissue between them. What if those agents need to work together to troubleshoot a complex problem, like a sudden and unexpected shift in product demand?

If they’re siloed, they cannot pool their abilities and function as a whole greater than the sum of their parts. Not orchestrating agents is like hiring several subcontractors to build a house but restricting their tools and communication. The result is a poorly built house – or jumble of agents with poor performance.

Agents and silos

Agents can also be siloed from the technology that enterprises already have in place. Imagine an HR Agent tasked with orchestrating employee PTO – but unable to access certain calendar applications and documents.

Imagine an IT Agent tasked with troubleshooting software problems – but unable to access troves of past incident reports and help desk tickets. These agents would fail to complete their fundamental tasks, and the time and resources that went into building them would be wasted.

There is something deeply ironic about siloed agents. Agents' value lies in their very ability to traverse the full enterprise stack, bridging tools and processes that require human time and talent. When agents get stuck, they are a victim of the very problem they are trying to solve. Businesses are investing in the problem, not the solution.

Siloed agents have an additional pitfall: they need to be governed and secured piecemeal. Relying on an ad hoc, patchwork approach to governance and security means an agentic use case is likely to fall through the cracks. If this occurs, agents’ most valuable asset – their autonomy – can quickly turn into a liability. Issues like bias, drift, and security vulnerabilities are amplified by agents’ access and independence.

Reaching potential

For agents to reach their full potential, business leaders must first fix the fragmentation underneath. Enterprises need a single data fabric that can unify the structured and unstructured data that powers agents. While many enterprises haven’t achieved this yet, a growing number understand the value: 72% of leaders view their organization’s proprietary data as key to unlocking the value of generative AI, according to my company’s most recent CEO Study.

Enterprises also need a hybrid control plane automating the sprawling landscape agents work across, unifying APIs, apps, events, files, and mainframe data. And enterprises should invest in a central nervous system for their agents. The future is multi-agent: It will be teams of agents, rather than a single agent, that tackle complex tasks. Enterprises need a single hub to supervise and route those agents. In other words, enterprises need a general contractor for all those subcontractors.

The need for orchestration

Better integrated and orchestrated agents also boost observability. Rather than governing and securing agents piecemeal, enterprises can apply comprehensive rules and oversight from a single point. This also allows AI security teams and AI governance teams to collaborate: if a shadow agent deployment is spotted by security tools, it can quickly and automatically steer the agent into the proper governance workstream.

Enterprises are rightfully investing in agents. But if they want that investment to translate into impact, they should be making equal commitments to agent integration and orchestration. Otherwise, they will end up with a whole that is less than the sum of its parts. In 2025 and beyond, it won't just be the businesses with the best agents that win. It will be the businesses with the most flexible ones.

I tried 70+ best AI tools.

This article was produced as part of TechRadarPro's Expert Insights channel where we feature the best and brightest minds in the technology industry today. The views expressed here are those of the author and are not necessarily those of TechRadarPro or Future plc. If you are interested in contributing find out more here: https://www.techradar.com/news/submit-your-story-to-techradar-pro

• Ryzen Threadripper Pro 9995WX now on sale with Zen 5, 96 cores, 192 threads
• AMD’s most powerful workstation CPU available from major retailers for $11,699
• Most professionals won’t need the 9995WX despite the impressive specs

The AMD Ryzen Threadripper Pro 9995WX is now available to buy from major retailers, including Amazon and Newegg, with a starting price of $11,699 - much cheaper than initial predictions.

It’s the top chip in AMD’s new Threadripper Pro 9000 WX-series and is built on the latest Zen 5 architecture, sporting 96 cores, 192 threads, and offers a boost clock of up to 5.4GHz.

The 9995WX uses TSMC’s 4nm process and fits the sTR5 socket. The CPU supports 8-channel DDR5-6400 memory, offers up to 148 PCIe lanes, and carries a 350W TDP.

For professional workloads

This processor is aimed at professional workloads that can scale across dozens of threads.

AMD lists compatibility with chipsets like Pro 695, TRX50, and WRX90.

The chip also supports ECC memory and includes features like AMD Pro technologies and EXPO for memory tuning.

For those running highly parallel tasks like visual effects rendering or scientific modeling, this processor can deliver impressive throughput, but at this price, it’s hard to recommend unless the application is truly pushing the limits of core count.

While AMD’s new CPU can deliver unmatched multi-threaded performance in the right environment, most professionals will likely find better value in AMD's lower-tier options (the EPYC 9655 matches Threadripper PRO 9995WX core count for less than half the price) or in Intel's workstation CPUs, depending on their specific use case.

With no integrated graphics, a discrete GPU is required. Cooling is also left to the buyer, as no cooler is bundled.

Unless you're operating in a niche that truly benefits from 96 cores, it’s unlikely to be the best choice or offer the best value for most users.

Buyers in the UK can, as always, expect to pay slightly more than their American counterparts. Novatech is currently selling the new chip there for £10,510, or around $14,205.

You may also like

• 16-Core AMD EPYC 4005 CPU is almost 3X faster than AMD's first server flagship
• Rare AMD CPU powers this just-announced NAS offering
• AMD just launched its 96-core Ryzen Threadripper Pro 9995WX CPU

A landlord sentenced to decades in prison after he killed a Palestinian American boy and wounded his mother has died.

(Image credit: Charles Rex Arbogast)

The American Medical Association is urging HHS Secretary Robert F. Kennedy Jr. not to oust members of the U.S. Preventive Services Task Force, an independent group of experts focused on primary care.

(Image credit: Brendan Smialowski)

The songwriter and math professor found a following with his satirical political songs.

(Image credit: Evening Standard)

Here are the answers for The New York Times Mini Crossword for July 28.

Ichiro Suzuki has just been inducted into the Baseball Hall of Fame. NPR's Scott Detrow talks to Shannon Drayer about one of the most fascinating careers in Major League Baseball history.

Some USA swim team members have missed events during the competition.

(Image credit: Vincent Thian)

Here are hints and the answers for the NYT Connections: Sports Edition puzzle for July 28, No. 308.

Here are some hints and the answers for the NYT Connections puzzle for July 28, #778.

Here are hints and answers for the NYT Strands puzzle for July 28 No. 512.

Here are hints and the answer for today's Wordle for July 28, No. 1,500.