DD
MM
YYYY

CATEGORIES

PAGES

DD
MM
YYYY

spot_img

CATEGORIES

PAGES

Home Blog Page 54

How AI Data Centers Are Making Everything More Expensive

By
Paul Ugbede Godwin
-
0

The artificial intelligence boom is transforming the global economy, but its costs are increasingly extending far beyond the technology industry.

Behind every AI chatbot, image generator and automated agent sits a rapidly expanding network of data centers packed with powerful chips.

These facilities require enormous amounts of electricity, water, land and construction materials, creating a new source of economic pressure that consumers are beginning to feel.

The biggest pressure point is electricity. AI data centers consume vastly more power than conventional computing facilities because advanced AI models require thousands of graphics processors and specialized accelerators operating around the clock.

As technology companies race to build larger facilities, demand for electricity is growing faster in some regions than utilities and grids can comfortably accommodate. That can push utilities to invest billions in new generation, transmission lines and substations.

Ultimately, those costs can find their way into electricity bills. The problem is particularly significant because AI companies are competing for the same limited power resources as households, manufacturers and other businesses.

In areas experiencing rapid data-center expansion, utilities may need to build infrastructure specifically to support large computing campuses. Even when technology companies contribute to those projects, some costs can be distributed across the wider electricity system.

Then there is the physical construction boom. Modern AI data centers are enormous industrial projects requiring concrete, steel, copper, electrical equipment, cooling systems and backup generators.

A surge in demand for these materials can increase prices, especially when supply chains are already constrained. Copper is particularly important because data centers require extensive electrical infrastructure.

While transformers and other specialized equipment can have lengthy manufacturing lead times.

Land is another increasingly valuable resource. Technology companies are seeking large parcels close to power generation and major transmission networks.

As competition intensifies, land values can rise, particularly around communities that become attractive destinations for data-center development. Housing costs can also come under pressure if thousands of construction workers, engineers and other employees move into smaller communities.

Water presents another concern. Many data centers require sophisticated cooling systems to prevent computing equipment from overheating. In water-stressed regions, competition between data centers, agriculture, households and other industries could become increasingly contentious.

Even when companies use more efficient cooling technologies, the scale of new facilities means their aggregate resource consumption can still be significant.

There is also a financial cost. AI infrastructure requires extraordinary amounts of capital, and companies are borrowing, raising money and entering massive infrastructure agreements to finance the buildout.

Investors may eventually demand higher returns to compensate for the enormous spending and operational risks. Those costs can influence the prices businesses charge for AI services. Yet the story is not entirely negative.

AI data centers can generate construction jobs, attract investment, strengthen local infrastructure and stimulate demand for new energy projects. The technologies developed to power them could accelerate renewable energy deployment, battery storage and grid modernization.

The central question is therefore not whether AI should expand, but who should pay for its expansion. If the benefits of artificial intelligence are captured mainly by technology companies while electricity, water, infrastructure and housing costs are distributed among ordinary consumers, the economic bargain becomes harder to justify.

AI may make businesses more productive and services cheaper. But during the infrastructure race required to build it, the opposite can happen. The hidden price of artificial intelligence is increasingly appearing in the cost of power, materials, land and public infrastructure.

The AI revolution may promise a cheaper future, but getting there could make the present considerably more expensive.

AI Moves From Commands to Continuous Interaction

By
Paul Ugbede Godwin
-
0

Artificial intelligence is entering a new phase in which the defining feature is no longer simply how intelligently a model can answer a question, but how effectively it can operate inside an ongoing workflow.

Two recent developments capture this transition: Nous Research has brought its Hermes Agent into an official desktop application, while Visko has demonstrated Orbis 1.0, a live video model designed to generate and modify virtual worlds in real time.

Nous Research’s Hermes Desktop gives its open-source Hermes Agent a native graphical interface for macOS, Windows and Linux. Previously, Hermes was primarily experienced through command-line tools and messaging interfaces.

The desktop application lowers that barrier by allowing users to interact with the agent through a conventional interface while retaining capabilities such as streaming tool output, file browsing, previews, voice interaction and persistent agent memory.

The significance of Hermes Desktop extends beyond convenience. Hermes is designed as a persistent, self-hosted agent rather than merely a chatbot.

Its architecture allows skills, memory, sessions and configurations to carry across interfaces. That means the desktop application can become an operating environment for an agent that performs tasks rather than simply responding to isolated prompts.

This reflects a broader industry movement toward agentic computing. Instead of asking an AI to generate an answer and then manually executing the result, users increasingly expect software to research information, manipulate files, use tools and maintain context across multiple interactions.

Hermes Desktop makes that model accessible without requiring users to live inside a terminal. At the same time, Visko is pushing the frontier from static video generation toward continuously evolving visual environments.

Its Orbis 1.0 model is described as a “Live Model” capable of generating interactive long-form video while allowing users to change prompts during generation. The system supports text-to-video, image-to-video and video continuation, while maintaining visual consistency across extended sequences.

The technical claims are particularly ambitious. Visko says Orbis can generate 4K video at 24 frames per second in real time and sustain hour-scale generation without obvious visual or color drift.

Its architecture combines a streaming generator, video upscaler and bounded multi-scale memory designed to preserve subjects, scenes and styles as the world continues evolving.

The most important distinction is interactivity. Traditional video models generally operate as a request-and-response system: enter a prompt, wait for a clip, then start again. Orbis instead treats generation as an ongoing process.

Users can alter the narrative while the video is running, effectively steering the world rather than repeatedly regenerating disconnected clips. Visko’s demonstration includes live prompt changes and voice-driven interaction.

Hermes Desktop and Orbis 1.0 point toward a common direction for AI: continuous computing. Hermes keeps an agent active across tasks and sessions; Orbis keeps a generated world active across time. One turns software into an ongoing collaborator, while the other turns video generation into an ongoing environment.

The implications could extend across entertainment, gaming, simulation, education, robotics, research and digital production. If these systems mature, interacting with AI may become less like querying a database and more like entering a persistent digital space where agents remember, act and adapt.

The next major AI competition may therefore not be about who produces the best single response. It may be about who can build the most capable systems that keep going after the prompt ends.

Chinese Bots and the Growing US Data-Center Backlash

By
Paul Ugbede Godwin
-
0

The debate over artificial intelligence infrastructure is taking an unexpected turn: the fight over data centers is no longer confined to electricity grids, water consumption, land use, and local politics. It is increasingly becoming a battle over information itself.

A recent investigation has raised questions about claims that Chinese-linked bots on X were amplifying anti-data-center narratives in the United States. According to a researcher examining the accounts identified by X.

Many appeared to have virtually no audience or meaningful engagement. Some reportedly had no followers at all, making their ability to influence public opinion difficult to establish.

That distinction matters. A social-media account can publish hundreds of posts, but if nobody sees, shares, replies to, or interacts with them, its practical influence may be negligible.

The existence of automated or coordinated accounts does not automatically prove that they successfully shaped public sentiment.

The controversy comes at a sensitive moment. Data centers have become one of the most politically contentious parts of America’s AI boom. Companies developing artificial intelligence are spending billions of dollars on computing infrastructure.

While communities are increasingly questioning who pays for the electricity, water, roads, and other infrastructure required to support massive facilities. Opposition has emerged from both sides of the political spectrum.

Residents have complained about noise, land use, energy demand and environmental consequences. Utilities are also facing difficult questions about how quickly electricity generation and transmission capacity can expand to meet surging demand from AI companies.

That genuine grassroots opposition creates an important backdrop for the bot controversy. Even if foreign actors attempted to amplify criticism, it would not necessarily mean that the underlying concerns were manufactured.

This is where social-media attribution becomes complicated. Researchers and platforms can identify suspicious accounts using signals such as coordinated posting patterns, account creation dates, language behavior and network relationships.

But determining whether those accounts actually changed people’s opinions is a much harder task. An account with zero followers can still contribute to a coordinated influence operation if its content is later amplified by larger accounts.

Conversely, a large collection of seemingly suspicious accounts may have almost no measurable impact if their posts remain isolated. The episode therefore highlights a broader problem confronting X and other social platforms: distinguishing between inauthentic activity and actual influence.

Platforms have strong incentives to expose foreign influence campaigns, particularly when they involve geopolitical rivals such as China. But credibility depends on providing enough evidence to demonstrate not only that suspicious accounts existed.

But also how they operated and what impact they had. For policymakers, the distinction is equally important. If legitimate opposition to data centers is dismissed as foreign propaganda, communities with genuine concerns may feel ignored.

At the same time, policymakers cannot afford to overlook coordinated foreign attempts to manipulate political debates surrounding strategically important technologies.

The AI infrastructure race is already producing enormous economic and geopolitical consequences. Data centers are becoming critical national infrastructure, while access to advanced computing is increasingly viewed as an element of national power.

That makes the information war surrounding them inevitable. But the lesson from the latest bot investigation should be caution rather than panic. Suspicious accounts are evidence of suspicious activity—not automatically evidence of successful influence.

The real question is whether those accounts reached real audiences, changed perceptions, and materially affected the political debate. As America’s data-center expansion accelerates, separating authentic public opposition from artificial amplification will become increasingly important.

The future of AI infrastructure may depend not only on who builds the biggest computers, but also on who controls the narrative surrounding them.

Elon Musk’s Grok 4.7 Bets on Scale, Data and Real-World Engineering

By
Paul Ugbede Godwin
-
0

Elon Musk is once again raising the stakes in the artificial intelligence race, claiming that xAI’s next major model, Grok 4.7, could be released within 10 days and outperform every AI model currently available.

The prediction reflects Musk’s increasingly aggressive push to position Grok alongside, and potentially ahead of, the leading systems developed by OpenAI and Anthropic.

The reported upgrade is significant in both scale and ambition. Grok 4.7 is expected to expand from approximately 1.5 trillion parameters to 2.1 trillion, representing a substantial increase in model capacity.

While parameter count alone does not determine an AI system’s intelligence, the expansion signals xAI’s willingness to continue competing through massive computational resources, increasingly sophisticated training techniques and access to specialized data.

Perhaps more important than the additional parameters is the reported use of proprietary SpaceX data. Musk argues that information generated through SpaceX’s engineering and technological operations could give Grok an advantage on real-world technical problems.

Such data could potentially expose the model to highly specialized engineering knowledge, systems analysis and problem-solving scenarios that are difficult to obtain through conventional internet-scale training.

That strategy could distinguish Grok from competitors whose training datasets are primarily composed of publicly available information, licensed material and synthetic data.

If successfully integrated, proprietary industrial data could make AI models more useful for advanced engineering, scientific research and technical decision-making.

However, the quality, relevance and deployment of the data will matter more than simply possessing a larger dataset. Musk has also identified Anthropic as Grok’s closest competitor, while acknowledging the company’s ability to develop increasingly capable models.

That assessment highlights how competitive the frontier AI market has become. OpenAI, Anthropic and xAI are no longer simply competing over chatbot quality. They are racing across coding, reasoning, autonomous agents, scientific discovery, computer use and enterprise applications.

Grok’s existing benchmark performance illustrates why Grok 4.7 faces a difficult test. Grok 4.6 reportedly matched GPT-5.6 Sol Max on the AA Intelligence Index but remained behind Claude Fable 5 Max. Its reported 26% score on Terminal-Bench also trailed GPT’s 34.6%.

These results suggest that although Grok is firmly among the leading AI systems, there remains a measurable gap in certain forms of complex reasoning and agentic computer-based work. That makes Grok 4.7’s promised improvement particularly significant.

If the new model genuinely delivers a major leap in coding, reasoning and technical problem-solving, xAI could challenge the established hierarchy of frontier models. Conversely, if benchmark improvements are modest, Musk’s claim that Grok will outperform every available model could prove overly ambitious.

The distinction between marketing claims and demonstrated capability will therefore be crucial. Frontier AI development has become increasingly competitive, and companies routinely make bold predictions before independent evaluations become available.

Objective benchmarks, real-world testing and user experience will ultimately determine whether Grok 4.7 represents a genuine technological breakthrough.

For xAI, the strategy is clear: combine enormous model scale with proprietary technical data and Musk’s broader technology ecosystem. The coming release could provide an important test of whether access to unique engineering information can translate into superior artificial intelligence.

When independent results arrive, the industry will have a clearer answer to the central question: can Grok 4.7 turn Musk’s boldest AI prediction into measurable performance?

1...535455...9,314Page 54 of 9,314

© Tekedia. All rights reserved.

Term & Privacy

© Tekedia. All rights reserved.

Term & Privacy