#:is

New Post has been published on https://thedigitalinsider.com/why-genai-stalls-without-strong-governance/

Why GenAI Stalls Without Strong Governance

As companies grapple with moving Generative AI projects from experimentation to productionising – many businesses remain stuck in pilot mode. As our recent research highlights, 92% of organisations are concerned that GenAI pilots are accelerating without first tackling fundamental data issues. Even more telling: 67% have been unable to scale even half of their pilots to production. This production gap is less about technological maturity and more about the readiness of the underlying data. The potential of GenAI depends upon the strength of the ground it stands on. And today, for most organisations, that ground is shaky at best.

Why GenAI gets stuck in pilot

Although GenAI solutions are certainly mighty, they’re only as effective as the data that feeds them. The old adage of “garbage in, garbage out” is truer today than ever. Without trusted, complete, entitled and explainable data, GenAI models often produce results that are inaccurate, biased, or unfit for purpose.

Unfortunately, organisations have rushed to deploy low-effort use cases, like AI-powered chatbots offering tailored answers from different internal documents. And while these do improve customer experiences to an extent, they don’t demand deep changes to a company’s data infrastructure. But to scale GenAI strategically, whether in healthcare, financial services, or supply chain automation, requires a different level of data maturity.

In fact, 56% of Chief Data Officers cite data reliability as a key barrier to the deployment of AI. Other issues are incomplete data (53%), privacy issues (50%), and larger AI governance gaps (36%).

No governance, no GenAI

To take GenAI beyond the pilot stage, companies must treat data governance as a strategic imperative to their business.They need to ensure data is up to the job of powering AI models, and to so the following questions need to be addressed:

Is the data used to train the model coming from the right systems?

Have we removed personally identifiable information and followed all data and privacy regulations?

Are we transparent, and can we prove the lineage of the data the model uses?

Can we document our data processes and be ready to show that the data has no bias?

Data governance also needs to be embedded within an organisation’s culture. To do this, requires building AI literacy across all teams. The EU AI Act formalises this responsibility, requiring both providers and users of AI systems to make best efforts to ensure employees are sufficiently AI-literate, making sure they understand how these systems work and how to use them responsibly. However, effective AI adoption goes beyond technical know-how. It also demands a strong foundation in data skills, from understanding data governance to framing analytical questions. Treating AI literacy in isolation from data literacy would be short-sighted, given how closely they’re intertwined.

In terms of data governance, there’s still work to be done. Among businesses who want to increase their data management investments, 47% agree that lack of data literacy is a top barrier. This highlights the need for building top-level support and developing the right skills across the organisation is crucial. Without these foundations, even the most powerful LLMs will struggle to deliver.

Developing AI that must be held accountable

In the current regulatory environment, it’s no longer enough for AI to “just work,” it also needs to be accountable and explained. The EU AI Act and the UK’s proposed AI Action Plan requires transparency in high-risk AI use cases. Others are following suit, and 1,000+ related policy bills are on the agenda in 69 countries.

This global movement towards accountability is a direct result of increasing consumer and stakeholder demands for fairness in algorithms. For example, organisations must be able to say the reasons why a customer was turned down for a loan or charged a premium insurance rate. To be able to do that, they would need to know how the model made that decision, and that in turn hinges on having a clear, auditable trail of the data that was used to train it.

Unless there is explainability, businesses risk losing customer trust as well as facing financial and legal repercussions. As a result, traceability of data lineage and justification of results is not a “nice to have,” but a compliance requirement.

And as GenAI expands beyond being used for simple tools to fully-fledged agents that can make decisions and act upon them, the stakes for strong data governance rise even higher.

Steps for building trustworthy AI

So, what does good look like? To scale GenAI responsibly, organisations should look to adopt a single data strategy across three pillars:

Tailor AI to business: Catalogue your data around key business objectives, ensuring it reflects the unique context, challenges, and opportunities specific to your business.

Establish trust in AI: Establish policies, standards, and processes for compliance and oversight of ethical and responsible AI deployment.

Build AI data-ready pipelines: Combine your diverse data sources into a resilient data foundation for robust AI baking in prebuilt GenAI connectivity.

When organisations get this right, governance accelerates AI value. In financial services for example, hedge funds are using gen AI to outperform human analysts in stock price prediction while significantly reducing costs. In manufacturing, supply chain optimisation driven by AI enables organisations to react in real-time to geopolitical changes and environmental pressures.

And these aren’t just futuristic ideas, they’re happening now, driven by trusted data.

With strong data foundations, companies reduce model drift, limit retraining cycles, and increase speed to value. That’s why governance isn’t a roadblock; it’s an enabler of innovation.

What’s next?

After experimentation, organisations are moving beyond chatbots and investing in transformational capabilities. From personalising customer interactions to accelerating medical research, improving mental health and simplifying regulatory processes, GenAI is beginning to demonstrate its potential across industries.

New Post has been published on https://thedigitalinsider.com/deepseek-distractions-why-ai-native-infrastructure-not-models-will-define-enterprise-success/

DeepSeek Distractions: Why AI-Native Infrastructure, Not Models, Will Define Enterprise Success

Imagine trying to drive a Ferrari on crumbling roads. No matter how fast the car is, its full potential is wasted without a solid foundation to support it. That analogy sums up  today’s enterprise AI landscape. Businesses often obsess over shiny new models like DeepSeek-R1 or OpenAI o1 while neglecting the importance of infrastructure to derive value from them. Instead of solely focusing on who’s building the most advanced models, businesses need to start investing in robust, flexible, and secure infrastructure that enables them to work effectively with any AI model, adapt to technological advancements, and safeguard their data.

With the release of DeepSeek, a highly sophisticated large language model (LLM) with controversial origins, the industry is currently gripped by two questions:

Is DeepSeek real or just smoke and mirrors?

Did we over-invest in companies like OpenAI and NVIDIA?

Tongue-in-cheek Twitter comments imply that DeepSeek does what Chinese technology does best: “almost as good, but way cheaper.” Others imply that it seems too good to be true. A month after its release, NVIDIA’s market dropped nearly $600 Billion and Axios suggests this could be an extinction-level event for venture capital firms. Major voices are questioning whether Project Stargate’s $500 Billion commitment towards physical AI infrastructure investment is needed, just 7 days after its announcement.

And today, Alibaba just announced a model that claims to surpass DeepSeek!

AI models are just one part of the equation. It’s the shiny new object, not the whole package for Enterprises. What’s missing is AI-native infrastructure.

A foundational model is merely a technology—it needs capable, AI-native tooling to transform into a powerful business asset. As AI evolves at lightning speed, a model you adopt today might be obsolete tomorrow. What businesses really need is not just the “best” or “newest” AI model—but the tools and infrastructure to seamlessly adapt to new models and use them effectively.

Whether DeepSeek represents disruptive innovation or exaggerated hype isn’t the real question. Instead, organizations should set their skepticism aside and ask themselves if they  have the right AI infrastructure to stay resilient as models improve and change. And can they switch between models easily to achieve their business goals without reengineering everything?

Models vs. Infrastructure vs. Applications

To better understand the role of infrastructure, consider the three components of leveraging AI:

The Models: These are your AI engines—Large Language Models (LLMs) like ChatGPT, Gemini, and DeepSeek. They perform tasks such as language understanding, data classification, predictions, and more.

The Infrastructure: This is the foundation on which AI models operate. It includes the tools, technology, and managed services necessary to integrate, manage, and scale models while aligning them with business needs. This generally includes technology that focuses on Compute, Data, Orchestration and Integration. Companies like Amazon and Google provide the infrastructure to run models, and tools to integrate them into an enterprise’s tech stack.

The Applications/Use Cases: These are the apps that end users see that utilize AI models to accomplish a business outcome. Hundreds of offerings are entering the market from incumbents bolting on AI to existing apps (i.e., Adobe, Microsoft Office with Copilot.) and their AI-native challengers (Numeric, Clay, Captions).

While models and applications often steal the spotlight, infrastructure quietly enables everything to work together smoothly and sets the foundation for how models and applications operate in the future. It ensures organizations can switch between models and unlock the real value of AI—without breaking the bank or disrupting operations.

Why AI-native infrastructure is mission-critical

Each LLM excels at different tasks. For example, ChatGPT is great for conversational AI, while Med-PaLM is designed to answer medical questions. The landscape of AI is so hotly contested that today’s top-performing model could be eclipsed by a cheaper, better competitor tomorrow.

Without flexible infrastructure, companies may find themselves locked into one model, unable to switch without completely rebuilding their tech stack. That’s a costly and inefficient position to be in. By investing in infrastructure that is model-agnostic, businesses can integrate the best tools for their needs—whether it’s transitioning from ChatGPT to DeepSeek, or adopting an entirely new model that launches next month.

An AI model that is cutting-edge today may become obsolete in weeks. Consider hardware advancements like GPUs—businesses wouldn’t replace their entire computing system for the newest GPU; instead, they’d ensure their systems can adapt to newer GPUs seamlessly. AI models require the same adaptability. Proper infrastructure ensures enterprises can consistently upgrade or switch their models without reengineering entire workflows.

Much of the current enterprise tooling is not built with AI in mind. Most data tools—like those that are part of the traditional analytics stack—are designed for code-heavy, manual data manipulation. Retrofitting AI into these existing tools often creates inefficiencies and limits the potential of advanced models.

AI-native tools, on the other hand, are purpose-built to interact seamlessly with AI models. They simplify processes, reduce reliance on technical users, and leverage AI’s ability to not just process data but extract actionable insights. AI-native solutions can abstract complex data and make it usable by AI for querying or visualization purposes.

Core pillars of AI infrastructure success

To future-proof your business, prioritize these foundational elements for AI infrastructure:

Data Abstraction Layer

Think of AI as a “super-powered toddler.” It’s highly capable but needs clear boundaries and guided access to your data. An AI-native data abstraction layer acts as a controlled gateway, ensuring your LLMs only access relevant information and follow proper security protocols. It can also enable consistent access to metadata and context no matter what models you are using.

Explainability and Trust

AI outputs can often feel like black boxes—useful, but hard to trust. For example, if your model summarizes six months of customer complaints, you need to understand not only how this conclusion was reached but also what specific data points informed this summary.

AI-native Infrastructure must include tools that provide explainability and reasoning—allowing humans to trace model outputs back to their sources, and understand the reason for the outputs. This enhances trust and ensures repeatable, consistent results.

Semantic Layer

A semantic layer organizes data so that both humans and AI can interact with it intuitively. It abstracts the technical complexity of raw data and presents meaningful business information as context to LLMs while answering business questions. A well nourished semantic layer can significantly reduce LLM hallucinations.  .

For instance, an LLM application with a powerful semantic layer could not only analyze your customer churn rate but also explain why customers are leaving, based on tagged sentiment in customer reviews.

Flexibility and Agility

Your infrastructure needs to enable agility—allowing organizations to switch models or tools based on evolving needs. Platforms with modular architectures or pipelines  can provide this agility. Such tools allow businesses to test and deploy multiple models simultaneously and then scale the solutions that demonstrate the best ROI.

Governance Layers for AI Accountability 

AI governance is the backbone of responsible AI use. Enterprises need robust governance layers to ensure models are used ethically, securely, and within regulatory guidelines. AI governance manages three things.

Access Controls: Who can use the model and what data can it access?

Transparency: How are outputs generated and can the AI’s recommendations be audited?

Risk Mitigation:Preventing AI from making unauthorized decisions or using sensitive data improperly.

Imagine a scenario where an open-source model like DeepSeek is given access to SharePoint document libraries . Without governance in place, DeepSeek can answer questions that could include sensitive company data, potentially leading to catastrophic breaches or misinformed analyses that damage the business. Governance layers reduce this risk, ensuring AI is deployed strategically and securely across the organization.

Why infrastructure is especially critical now

Let’s revisit DeepSeek. While its long-term impact remains uncertain, it’s clear that global AI competition is heating up. Companies operating in this space can no longer afford to rely on assumptions that one country, vendor, or technology will maintain dominance forever.

Without robust infrastructure:

Businesses are at greater risk of being stuck with outdated or inefficient models.

Transitioning between tools becomes a time-consuming, expensive process.

Teams lack the ability to audit, trust, and understand the outputs of AI systems clearly.

Infrastructure doesn’t just make AI adoption easier—it unlocks AI’s full potential.

Build roads instead of buying engines

Models like DeepSeek, ChatGPT, or Gemini might grab headlines, but they are only one piece of the larger AI puzzle. True enterprise success in this era depends on strong, future-proofed AI infrastructure that allows adaptability and scalability.

Don’t get distracted by the “Ferraris” of AI models. Focus on building the “roads”—the infrastructure—to ensure your company thrives now and in the future.

New Post has been published on https://thedigitalinsider.com/fancy-menu-navigation-using-anchor-positioning/

Fancy Menu Navigation Using Anchor Positioning

You have for sure heard about the new CSS Anchor Positioning, right? It’s a feature that allows you to link any element from the page to another one, i.e., the anchor. It’s useful for all the tooltip stuff, but it can also create a lot of other nice effects.

In this article, we will study menu navigation where I rely on anchor positioning to create a nice hover effect on links.

Cool, right? We have a sliding effect where the blue rectangle adjusts to fit perfectly with the text content over a nice transition. If you are new to anchor positioning, this example is perfect for you because it’s simple and allows you to discover the basics of this new feature. We will also study another example so stay until the end!

Note that only Chromium-based browsers fully support anchor positioning at the time I’m writing this. You’ll want to view the demos in a browser like Chrome or Edge until the feature is more widely supported in other browsers.

The initial configuration

Let’s start with the HTML structure which is nothing but a nav element containing an unordered list of links:

<nav> <ul> <li><a href="#">Home</a></li> <li class="active"><a href="#">About</a></li> <li><a href="#">Projects</a></li> <li><a href="#">Blog</a></li> <li><a href="#">Contact</a></li> </ul> </nav>

We will not spend too much time explaining this structure because it can be different if your use case is different. Simply ensure the semantic is relevant to what you are trying to do. As for the CSS part, we will start with some basic styling to create a horizontal menu navigation.

ul padding: 0; margin: 0; list-style: none; display: flex; gap: .5rem; font-size: 2.2rem; ul li a color: #000; text-decoration: none; font-weight: 900; line-height: 1.5; padding-inline: .2em; display: block;

Nothing fancy so far. We remove some default styling and use Flexbox to align the elements horizontally.

Sliding effect

First off, let’s understand how the effect works. At first glance, it looks like we have one rectangle that shrinks to a small height, moves to the hovered element, and then grows to full height. That’s the visual effect, but in reality, more than one element is involved!

Here is the first demo where I am using different colors to better see what is happening.

Each menu item has its own “element” that shrinks or grows. Then we have a common “element” (the one in red) that slides between the different menu items. The first effect is done using a background animation and the second one is where anchor positioning comes into play!

The background animation

We will animate the height of a CSS gradient for this first part:

/* 1 */ ul li background: conic-gradient(lightblue 0 0) bottom/100% 0% no-repeat; transition: .2s; /* 2 */ ul li:is(:hover,.active) background-size: 100% 100%; transition: .2s .2s; /* 3 */ ul:has(li:hover) li.active:not(:hover) background-size: 100% 0%; transition: .2s;

We’ve defined a gradient with a 100% width and 0% height, placed at the bottom. The gradient syntax may look strange, but it’s the shortest one that allows me to have a single-color gradient.

Related: “How to correctly define a one-color gradient”

Then, if the menu item is hovered or has the .active class, we make the height equal to 100%. Note the use of the delay here to make sure the growing happens after the shrinking.

Finally, we need to handle a special case with the .active item. If we hover any item (that is not the active one), then the .active item gets the shirking effect (the gradient height is equal to 0%). That’s the purpose of the third selector in the code.

Our first animation is done! Notice how the growing begins after the shrinking completes because of the delay we defined in the second selector.

The anchor positioning animation

The first animation was quite easy because each item had its own background animation, meaning we didn’t have to care about the text content since the background automatically fills the whole space.

We will use one element for the second animation that slides between all the menu items while adapting its width to fit the text of each item. This is where anchor positioning can help us.

Let’s start with the following code:

ul:before content:""; position: absolute; position-anchor: --li; background: red; transition: .2s; ul li:is(:hover, .active) anchor-name: --li; ul:has(li:hover) li.active:not(:hover) anchor-name: none;

To avoid adding an extra element, I will prefer using a pseudo-element on the ul. It should be absolutely-positioned and we will rely on two properties to activate the anchor positioning.

We define the anchor with the anchor-name property. When a menu item is hovered or has the .active class, it becomes the anchor element. We also have to remove the anchor from the .active item if another item is in a hovered state (hence, the last selector in the code). In other words, only one anchor is defined at a time.

Then we use the position-anchor property to link the pseudo-element to the anchor. Notice how both use the same notation --li. It’s similar to how, for example, we define @keyframes with a specific name and later use it inside an animation property. Keep in mind that you have to use the <dashed-indent> syntax, meaning the name must always start with two dashes (--).

The pseudo-element is correctly placed but nothing is visible because we didn’t define any dimension! Let’s add the following code:

ul:before bottom: anchor(bottom); left: anchor(left); right: anchor(right); height: .2em;

The height property is trivial but the anchor() is a newcomer. Here’s how Juan Diego describes it in the Almanac:

The CSS anchor() function takes an anchor element’s side and resolves to the <length> where it is positioned. It can only be used in inset properties (e.g. top, bottom, bottom, left, right, etc.), normally to place an absolute-positioned element relative to an anchor.

Let’s check the MDN page as well:

The anchor() CSS function can be used within an anchor-positioned element’s inset property values, returning a length value relative to the position of the edges of its associated anchor element.

Usually, we use left: 0 to place an absolute element at the left edge of its containing block (i.e., the nearest ancestor having position: relative). The left: anchor(left) will do the same but instead of the containing block, it will consider the associated anchor element.

That’s all — we are done! Hover the menu items in the below demo and see how the pseudo-element slides between them.

Each time you hover over a menu item it becomes the new anchor for the pseudo-element (the ul:before). This also means that the anchor(...) values will change creating the sliding effect! Let’s not forget the use of the transition which is important otherwise, we will have an abrupt change.

We can also write the code differently like this:

ul:before content:""; position: absolute; inset: auto anchor(right, --li) anchor(bottom, --li) anchor(left, --li); height: .2em; background: red; transition: .2s;

In other words, we can rely on the inset shorthand instead of using physical properties like left, right, and bottom, and instead of defining position-anchor, we can include the anchor’s name inside the anchor() function. We are repeating the same name three times which is probably not optimal here but in some situations, you may want your element to consider multiple anchors, and in such cases, this syntax will make sense.

Combining both effects

Now, we combine both effects and, tada, the illusion is perfect!

Pay attention to the transition values where the delay is important:

ul:before transition: .2s .2s; ul li transition: .2s; ul li:is(:hover,.active) transition: .2s .4s; ul:has(li:hover) li.active:not(:hover) transition: .2s;

We have a sequence of three animations — shrink the height of the gradient, slide the pseudo-element, and grow the height of the gradient — so we need to have delays between them to pull everything together. That’s why for the sliding of the pseudo-element we have a delay equal to the duration of one animation (transition: .2 .2s) and for the growing part the delay is equal to twice the duration (transition: .2s .4s).

Bouncy effect? Why not?!

Let’s try another fancy animation in which the highlight rectangle morphs into a small circle, jumps to the next item, and transforms back into a rectangle again!

I won’t explain too much for this example as it’s your homework to dissect the code! I’ll offer a few hints so you can unpack what’s happening.

Like the previous effect, we have a combination of two animations. For the first one, I will use the pseudo-element of each menu item where I will adjust the dimension and the border-radius to simulate the morphing. For the second animation, I will use the ul pseudo-element to create a small circle that I move between the menu items.

Here is another version of the demo with different coloration and a slower transition to better visualize each animation:

The tricky part is the jumping effect where I am using a strange cubic-bezier() but I have a detailed article where I explain the technique in my CSS-Tricks article “Advanced CSS Animation Using cubic-bezier()”.

Conclusion

I hope you enjoyed this little experimentation using the anchor positioning feature. We only looked at three properties/values but it’s enough to prepare you for this new feature. The anchor-name and position-anchor properties are the mandatory pieces for linking one element (often called a “target” element in this context) to another element (what we call an “anchor” element in this context). From there, you have the anchor() function to control the position.

New Post has been published on https://thedigitalinsider.com/ai-in-art-everything-you-should-know-about-its-role-and-future/

AI in Art: Everything You Should Know About Its Role and Future

There is a famous quote by Albert Einstein that says, “Creativity is intelligence having fun.” But what happens when intelligence is no longer just human?

Then, it becomes an artificial intelligence (AI). Read on to learn how AI in art is making waves and why it has become the talk of the town.

The following article will discuss AI’s role in art, the opportunities it offers, the challenges it presents, and more.

The Intersection of AI in Art and Creativity

AI has changed the world of art by boosting creativity, automating processes, and generating unique works. It collaborates with artists to generate new possibilities, blurring boundaries between human and machine creativity.

However, there are some questions:

Is AI truly creative, or is it mimicking human work?

How does it change our understanding of art?

Is AI just a copycat or a creative genius?

Opinions differ. Some critics argue that AI is merely a reflection of what already exists, copying patterns without any true originality. However, others see AI as a powerful creative tool that expands human imagination. As David Holtz, the founder of MidJourney, puts it, “We see this technology as an engine for the imagination.”

Applications of AI in Art

Here are a few applications of AI in art:

Generative AI

Generative AI is a new way of making art with the help of machines. Artists give the AI specific instructions, and it creates unique artwork based on those inputs. This can include anything from abstract paintings to digital sculptures.

Tools and Technologies Behind Gen AI in Art

Generative Adversarial Networks (GANs) are a key technology behind AI art. GANs use two neural networks working together. One network, the “generator,” creates images, while the other, the “discriminator,” checks if the images look real. This back-and-forth process helps the generator refine its output, producing more convincing and realistic images over time.

When AI in Art Gets Noticed

Artists like Memo Akten and Mike Tyka used the DeepDream algorithm to create art. They presented it at a gallery in San Francisco in 2016. This was one of the first times AI-generated art was displayed in a public exhibition.

The big moment came in 2018. The Portrait of Edmond Belamy, an AI-generated artwork created by the Paris-based group Obvious, was sold at Christie’s auction house. It is considered the first widely publicized sale of AI art.

Style Transfer

Style transfer uses AI to blend the style of one image with the content of another. This means you can turn a simple photograph into an artwork inspired by Van Gogh, Picasso, or any other iconic artist.

Apps like Prisma and DeepArt analyze the patterns, colors, and textures of a style image and apply them to your chosen photo.

Enhancing Artistic Processes

Tools like Adobe Sensei help with color correction, object removal, and image resizing. Now, artists have enough free time to focus on creativity rather than repetitive tasks.

AI can also generate unique patterns, music, and visuals that may be difficult for humans to imagine or produce. AI tends to be a modern-day muse that encourages artists to experiment and think outside the box.

Personalized Design

AI helps create designs that feel personal to each user. It adjusts interfaces based on user behavior, preferences, and demographics.

Predictive Design

AI can also predict what design trends or customer needs will emerge. It analyzes data to forecast what elements will catch on in the future. Predictive design ensures that what’s created today will still be relevant tomorrow.

The Benefits of AI in Art and Design

Here are a few benefits of AI brings to art and design:

Enhanced Creativity: AI acts as a helpful companion for artists, opening doors to fresh possibilities beyond traditional methods. It can inspire new design ideas, recommend color schemes, and even bring concepts to life with realistic 3D models.

Efficiency and Productivity: AI Automates time-consuming tasks like color correction and resizing. This frees up time for creatives to refine concepts and explore new ideas.

Accessibility: AI makes advanced creative tools available to everyone, regardless of skill level. It empowers individuals without formal training to produce professional-quality work.

Innovation: AI tools enable the creation of new art forms by blending different mediums and techniques. They encourage interdisciplinary collaborations, such as AI in bio.

Art, Ethics, and AI: Who Holds the Brush?

The rise of AI-generated art has opened up a pandora’s box of ethical and philosophical questions.

Who Owns the Art?

One of the biggest questions is ownership. If an artificial intelligence creates a piece of art, who owns it? Current legal frameworks often struggle to address these complexities, leaving creators and innovators navigating uncharted territory.

Where Does Creativity Come From?

AI in art raises questions about originality. While some view it as just another medium, others worry it might strip away the personal, emotional touch that makes human art unique.

Are Privacy and Bias at Stake?

Privacy and bias are major concerns. AI systems often use personal data or copyrighted material without permission. This can violate privacy and create biased, stereotypical art.

Will Technology Replace Human Artists?

As AI-generated art becomes more advanced, illustrators, designers, and even fine artists may find their jobs at risk. Companies may opt for cheaper, faster AI solutions instead of human labor.

How Should We Move Forward Then?

Clear guidelines on AI training, fair compensation for artists, and a commitment to preserving the irreplaceable value of human creativity must be established. This approach can ensure that AI complements rather than competes with human artistry.

Challenges in AI-Driven Art and Design

While AI is overhauling the world of art and design, there are still some challenges.

Technological Constraints: AI struggles to replicate the emotion and personal intent that give art its soul. Although it can mimic styles and patterns, it lacks human elements.

Risk of Over-Reliance: Powerful AI tools may lead artists to prioritize ease over originality. Excessive dependence on machines can result in generic art, lacking the distinctive touch of human creativity.

Environmental Concerns: Training large AI models requires a large amount of energy, which adds to environmental concerns.

Resistance from Traditional Artists: Some artists fear AI diminishes the value of human creativity and effort. There’s worry about the commodification of art when machines mass-produce pieces that mimic human styles.

The Future of AI in Art and Design

Though many challenges exist, AI is also opening doors in art and design that we didn’t know existed. Let’s look at the arena where AI is set to be truly creative.

Interactive Art with AR and VR: AI and AR/VR will change how people experience art by making it interactive.

AI as Curators in Museums and Galleries: Algorithms will help curate exhibits by analyzing visitor interests, historical relevance, and thematic elements.

Personalized Art and Design: AI can create unique designs as per individual tastes, like customized paintings, furniture, or fashion pieces. It considers personal elements such as favorite colors, music, or emotional state to craft deeply connected art.

Expanding AI’s Influence Beyond Traditional Art: AI will contribute to architectural design, industrial innovation, and artifact restoration. It can digitally rebuild damaged art and assist in preserving cultural heritage.

The Final Stroke: Pixels, Paint, and Possibilities in the AI-Art Revolution

Artificial intelligence-based Art is not just ARTificial anymore. It’s practically intelligent at this point.

Artists, designers, and creators are using AI to amplify their vision, making what’s possible even more exciting. The creative possibilities are endless, and we’re only scratching the surface.

In a future where AI is both the artist and the medium, the question won’t be “Can machines make art?” but rather, “What can’t they make?”.

New Post has been published on https://thedigitalinsider.com/four-ways-ai-helps-emerging-e-commerce-platforms-compete-with-major-game-distributors/

Four Ways AI Helps Emerging E-Commerce Platforms Compete with Major Game Distributors

Over the past 12 years, computer and video game distribution strategies have undergone a seismic shift. Sales of digital games exceeded those of physical copies for the first time in 2013, and the trend was further accelerated by the 2020 lockdowns. In Italy, for example, the first week of confinement led digital game downloads to skyrocket by 174.9%.

Looking ahead, the market is poised to continue growing, with Statista projecting it will grow at a CAGR of 5.76% between now and 2027, eventually reaching a market volume of $25.4 billion by the end of that year. 

Despite this, competition remains fierce. The digital games market is dominated by only a handful of platforms, and with 94% of spending taking place digitally, that leaves very little room for new entrants. Established players — such as Steam and Epic Games Store in the PC sphere — take advantage of this to impose hefty fees on publishers. 

For these major entities, integrating AI into their operations is second-nature. However, for smaller, emerging platforms, AI could be a game-changer — one that allows them to challenge the incumbent oligopoly. 

While replicating successful AI implementations requires careful consideration of platform-specific characteristics and operational contexts, here are four ways in which AI can help fledgling e-commerce companies compete with digital distribution giants. 

#1: Enhancing fraud detection

On gaming platforms, fraud happens at a much larger scale — and more recurrently — than in other e-commerce verticals. Given its capacity to process and analyze vast amounts of transaction data, AI’s algorithms can promptly identify suspicious patterns or anomalies. 

By scouring through extensive transaction databases, machine learning algorithms can adapt and recognize fraudulent operations, ranging from uncommon user behaviors to irregular payment schemes and purchases from atypical geographic regions. 

In traditional rule-based systems, some of these indicators might go unnoticed, hindering a company’s ability to detect fraud and exposing it to potential financial losses. 

At our company, by implementing AI-powered software — developed by a third party — we have prevented approximately 95% of fraudulent transactions. We also work hand-in-hand with technology. Once an operation is flagged as dubious, our manager personally reviews it. Digital game keys are not released to the buyer until the purchase has been manually approved by our manager. 

#2: Streamlining Customer Support Queries

In e-commerce, AI-powered chatbots are one of the most common applications of artificial intelligence. 

Since there are many solutions in the market already, chatbots are relatively easy to implement, even without historical data. Because they can learn from user interactions, chatbots yield results practically right away, and help companies reduce their need for customer support staff. 

Additionally, they free up time for the existing customer support agents. 

In our experience, most queries received — around 70% — are pretty simple and repetitive. Examples include:

Is the game available for purchase?

When can I receive the game key?

How do I activate my license key? 

What is the status of my order?

In 80% of these cases, our AI bots have been quite successful at helping our users without needing to transfer them to a live operator. Thus, we can say that our bots cover roughly 56% of our incoming support requests, liberating valuable resources that were previously poured into support staff so that we can use them elsewhere in the company to enhance our growth. 

#3: Identifying UX conversion-driving patterns

A common dilemma e-commerce-oriented business owners face is identifying those factors that successfully drive conversion and those that don’t. 

This is another area where AI can help, by collecting user data that pinpoints recurring behavioral patterns that either lead or deter conversions. Based on this data, companies can make UX-centered adjustments to their website. 

Additionally, AI can create customer segments that boost the effectiveness of marketing efforts. Since it can create user profiles across various dimensions, AI can uncover connections and group look-alike segments that might not be obvious through manual reviews. For example, customers who purchase GTA 5 may also be interested in games from a different genre that, in principle, bears no relation to GTA 5. 

To facilitate this, we have implemented a third-party AI personalization solution from Retail Rocket. By leveraging historical customer purchase data, this tool helps us accomplish several tasks, such as providing personalized product recommendations — both on our website and through email — and identifying relationships between products, enabling us to suggest complementary purchases. 

Additionally, we can also time our customers’ next potential purchase. This also improves our timing for marketing messages. All in all, we can proudly say that these efforts have bolstered our sales via marketing channels by approximately 15%. 

#4: Forecasting sales

Given the time-sensitive nature of the gaming industry — for instance, Steam imposes constraints on how many keys publishers can generate — effective forecasting is key. 

Here, we have implemented a straightforward AI model that is based on two primary methods: time series forecasting and regression analysis. 

By detecting patterns, the former helps us predict future sales figures and adapt to seasonality, which is an important factor in the gaming field. On the other hand, the latter assists our team in establishing relationships between sales data and other variables — demographics, pricing, product categories, and more. 

Since there are wide divergences in these parameters — for example, there are sports games released annually, such as those by EA Sports, and other strategy games that span across decades — getting these critical factors right is of paramount importance for accurate forecasting. 

We first started with this in the spring of 2024, so, as of now, our results are similar to what we were achieving without AI. However, we expect that as we further calibrate and refine our model, and accumulate more historical data, our accuracy will significantly improve over time. 

Final thoughts

In some fields, such as gaming, AI can become a democratizing factor — one that enables emerging, high-potential platforms to compete with established behemoths. 

Having said this, to fully realize its potential, it is not so much about simply integrating AI for the sake of it, but about doing it right. 

For smaller companies that cannot afford to maintain an in-house team of AI specialists, a viable solution is to utilize existing third-party software. Some of these ready-made solutions can be used by regular developers, even if they are not specialized in AI. 

My suggestion is that you do not transition all of your workload right away to AI. Instead, take a gradual approach. For example, ask AI to handle 10% of user queries, or to dynamically price 10% of your products. 

New Post has been published on https://thedigitalinsider.com/how-to-identify-deepfake-videos-like-a-fact-checker/

How to Identify Deepfake Videos Like a Fact-Checker

Deepfakes are synthetic media where an individual replaces a person’s likeness with someone else’s. They’re becoming more common online, often spreading misinformation around the world. While some may seem harmless, others can have malicious intent, making it important for individuals to discern the truth from digitally crafted false content.

Unfortunately, not everyone can access state-of-the-art software to identify deepfake videos. Here’s a look at how fact-checkers examine a video to determine its legitimacy and how you can use their strategies for yourself.

1. Examine the Context

Scrutinizing the context in which the video is presented is vital. This means looking at the background story, the setting and whether the video’s events align with what you know to be true. Deepfakes often slip here, presenting content that doesn’t hold up against real-world facts or timelines upon closer inspection.

One example involves a deepfake of Ukrainian President Volodymyr Zelensky. In March 2022, a deepfake video surfaced on social media where Zelensky appeared to be urging Ukrainian troops to lay down their arms and surrender to Russian forces. 

Upon closer examination, several contextual clues highlighted the video’s inauthenticity. The Ukrainian government’s official channels and Zelensky himself didn’t share this message. Also, the timing and circumstances didn’t align with known facts about Ukraine’s stance and military strategy. The video’s creation aimed to demoralize Ukrainian resistance and spread confusion among the international community supporting Ukraine.

2. Check the Source

When you come across a video online, check for its source. Understanding where a video comes from is crucial because hackers could use it against you to deploy a cyberattack. Recently, 75% of cybersecurity professionals reported a spike in cyberattacks, with 85% noting the use of generative AI by malicious individuals. 

This ties back to the rise of deepfake videos, and professionals are increasingly dealing with security incidents that AI-generated content is fueling. Verify the source by looking for where the video originated. A video originating from a dubious source could be part of a larger cyberattack strategy. 

Trusted sources are less likely to spread deepfake videos, making them a safer bet for reliable information. Always cross-check videos with reputable news outlets or official websites to ensure what you’re viewing is genuine.

3. Look for Inconsistencies in Facial Expressions

One of the telltale signs of a deepfake is the presence of inconsistencies in facial expressions. While deepfake technology has advanced, it often struggles with accurately mimicking the subtle and complex movements that occur naturally when a person talks or expresses emotions. You can spot these by looking out for the following inconsistencies:

Unnatural blinking: Humans blink in a regular, natural pattern. However, deepfakes may either under-represent blinking or overdo it. For instance, a deepfake could show a person talking for an extended period without blinking or blinking too rapidly.

Lip sync errors: When someone speaks in a video, their lip movement may be off. Watch closely to see if the lips match the audio. In some deepfakes, the mismatch is subtle but detectable when looking closely.

Facial expressions and emotions: Genuine human emotions are complex and reflected through facial movements. Deepfakes often fail to capture this, leading to stiff, exaggerated or not fully aligned expressions. For example, a deepfake video might show a person smiling or frowning with less nuance, or the emotional reaction may not match the context of the conversation.

4. Analyze the Audio

Audio can also give you clues into whether a video is real or fake. Deepfake technology attempts to mimic voices, but discrepancies often give them away. For instance, pay attention to the voice’s quality and characteristics. Deepfakes can sound robotic or flat in their speech, or they may lack the emotional inflections an actual human would exhibit naturally.

Background noise and sound quality can also provide clues. A sudden change could suggest that parts of the audio were altered or spliced together. Authentic videos typically remain consistent throughout the entirety.

5. Investigate Lighting and Shadows

Lighting and shadows play a large part in revealing a video’s authenticity. Deepfake technology often struggles with accurately replicating how light interacts with real-world objects, including people. Paying close attention to lighting and shadows can help you spot various items that indicate whether it’s a deepfake.

In authentic videos, the subject’s lighting and surroundings should be consistent. Deepfake videos may display irregularities, such as the face being lit differently from the background. If the video’s direction or source of light doesn’t make sense, it could be a sign of manipulation.

Secondly, shadows should behave according to the light sources in the scene. In deepfakes, shadows can appear at wrong angles or fail to correspond with other objects. Anomalies in shadow size, direction, and the presence or absence of expected shadows give you an overall idea.

6. Check for Emotional Manipulation

Deepfakes do more than create convincing falsehoods — people often design them to manipulate emotions and provoke reactions. A key aspect of identifying such content is to assess whether it aims to trigger an emotional response that could cloud rational judgment.

For instance, consider the incident where an AI-generated image of a bomb at the Pentagon circulated on Twitter X. Despite being completely fabricated, the image’s alarming nature caused it to go viral and trigger widespread panic. As a result, a $500 billion loss in the stock market occurred.

Deepfake videos can stir the same amount of panic, especially when AI is involved. While evaluating these videos, ask yourself:

Is the content trying to evoke a strong emotional response, such as fear, anger or shock? Authentic news sources aim to inform, not incite.

Does the content align with current events or known facts? Emotional manipulation often relies on disconnecting the audience from rational analysis.

Are reputable sources reporting the same story? The absence of corroboration from trusted news outlets can indicate the fabrication of emotionally charged content. 

7. Leverage Deepfake Detection Tools

As deepfakes become more sophisticated, relying solely on human observation to identify them can be challenging. Fortunately, deepfake detection tools that use advanced technology to distinguish between real and fake are available. 

These tools can analyze videos for inconsistencies and anomalies that may not be visible to the naked eye. They leverage AI and machine learning by utilizing speech watermarking as one method. These technologies are trained to recognize the watermark’s placement to determine if the audio was tampered with.

Microsoft developed a tool called Video Authenticator, which provides a confidence score indicating the likelihood of a deepfake. Similarly, startups and academic institutions continually develop and refine technologies to keep pace with evolving deepfakes.

Detecting Deepfakes Successfully