Global Tech Excellence Awards

Tuesday, April 8, 2025

Beyond acceleration: the rise of Agentic AI

We already find ourselves at an inflection point with AI. According to a recent study by McKinsey, we’ve reached the turning point where ‘businesses must look beyond automation and towards AI-driven reinvention’ to stay ahead of the competition. While the era of AI-driven acceleration isn’t over, a new phase has already begun – one that goes beyond making existing workflows more efficient and moves toward replacing existing workflows and/or creating new ones.

This is the age of Agentic AI.

Truly autonomous AI agents are capable of reshaping operations entirely. Systems can act autonomously, make decisions, and adapt dynamically. These agents will go beyond conversational interfaces, responding to user input and proactively managing tasks, navigating complex IT environments, and orchestrating business processes.

However, this shift isn’t just about technology — it also comes with a few considerations. Companies will need to address regulatory challenges, build AI literacy, and focus on applied use cases with clear ROI if the evolution is to succeed.

Moving from acceleration to transformation

So far, companies have primarily used AI to accelerate existing processes, whether through chatbots improving customer interactions or AI-driven analytics optimising workflows. In the end, these implementations make businesses more efficient.

But acceleration alone is no longer enough to stay ahead in the game. The real opportunity lies in replacing outdated workflows entirely and creating new, previously impossible capabilities.

For example, AI plays a vital role in automating troubleshooting and enhancing security within the network industry. But what if AI could autonomously anticipate and predict failures, reconfigure networks proactively to avoid service level degradations in real time, and optimise performance without human intervention? As AI becomes more autonomous, its ability to not just assist but act independently will be key to unlocking new levels of productivity and innovation.

That’s what Agentic AI is about.

Navigating the AI regulatory landscape

However, as AI becomes more autonomous, the regulatory landscape governing its deployment will evolve in parallel. The introduction of the EU AI Act, alongside global regulatory frameworks, means companies must already navigate new compliance requirements related to AI transparency, bias mitigation, and ethical deployment.

That means AI governance can no longer be an afterthought.

AI-powered systems must be designed with built-in compliance mechanisms, data privacy protections, and explainability features to build trust among users and regulators alike. Zero-trust security models will also be crucial in mitigating risks, enforcing strict access controls, and ensuring that AI decisions remain auditable and secure.

The importance of AI literacy

As stated, the success of Agentic AI’s era will depend on more than just technical capabilities – it will require alignment between leadership, developers, and end-users. As AI becomes more advanced, AI literacy becomes a key differentiator, and companies must invest in upskilling their workforce to understand AI’s capabilities, limitations, and ethical considerations. A recent report by the ICT Workforce Consortium found that 92% of information and communication technology jobs are expected to undergo significant transformation due to advancements in AI. So, without proper AI education, businesses risk misalignment between AI implementers and those who use the technology.

This can lead to a lack of trust, slow adoption, and ineffective deployment, which can impact the bottom line. So, to unlock the full potential of Agentic AI, it’s essential to build AI literacy across all levels of the organisation.

As this new era of AI blooms, companies must learn from the current era of AI adoption: focus on applied use cases with tangible ROI. The days of experimenting with AI for innovation’s sake are ending – the next generation of AI deployments must prove their worth.

In networking, it could be projects such as AI-powered autonomous network optimisation. These systems do more than automate tasks; they continuously monitor network traffic, predict congestion points, and autonomously adjust configurations to ensure optimal performance. By providing proactive insights and real-time adjustments, these AI-driven solutions help companies prevent issues and outages before they occur.

This level of AI autonomy reduces human intervention and enhances overall security and operational efficiency.

Identifying and implementing high-value, high-impact Agentic AI use cases such as these will be vital.

Trust as the adoption hurdle

While we’re entering a new era, trust plays a key role in widespread AI adoption. Users must feel confident that AI decisions are accurate, fair, and explainable. Even the most advanced AI models will face challenges gaining acceptance without transparency.

This is particularly relevant as AI transitions from assisting users to making autonomous decisions. Whether AI agents manage IT infrastructure or drive customer interactions, organisations must ensure that AI decisions are auditable, unbiased, and aligned with business objectives.

Without transparency and accountability, companies may face resistance from both employees and customers.

The future of AI

Looking ahead, 2025 holds exciting potential for AI. As it reaches a new level of maturity, its success will depend on how well organisations, governments, and individuals adapt to its growing presence in everyday life. Moving beyond efficiency and automation, AI has the opportunity to become a powerful driver of intelligent decision-making, problem-solving, and innovation.

Organisations that harness Agentic AI effectively – balancing autonomy with oversight – will see the greatest benefits. However, success will require a commitment to transparency, education, and ethical deployment to build trust and ensure AI is a true enabler of progress.

Because AI is no longer just an accelerant, it is a transformative force reshaping how we work, communicate, and interact with technology.

Website: International Research Awards on Computer Vision #computervision #deeplearning #machinelearning #artificialintelligence #neuralnetworks, #imageprocessing #objectdetection #imagerecognition #faceRecognition #augmentedreality #robotics #techtrends #3Dvision #professor #doctor #institute #sciencefather #researchawards #machinevision #visiontechnology #smartvision #patternrecognition #imageanalysis #semanticsegmentation #visualcomputing #datascience #techinnovation #university #lecture #biomedical

Visit Our Website : computer.scifat.com Nomination Link : computer-vision-conferences.scifat.com/award-nomination Registration Link : computer-vision-conferences.scifat.com/award-registration Member Link : computer-vision-conferences.scifat.com/conference-membership/? ecategory=Membership&rcategory=Member

Awards-Winners : computer-vision-conferences.scifat.com/awards-winners

Saturday, April 5, 2025

Machine learning models predict dementia risk among American Indian/Alaska Native adults

Machine learning algorithms utilizing electronic health records can effectively predict two-year dementia risk among American Indian/Alaska Native adults aged 65 years and older, according to a University of California, Irvine-led study. The findings provide a valuable framework for other healthcare systems, particularly those serving resource-limited populations.

The computer modeling results also found several new predictors for dementia diagnosis that were identified consistently across different machine-learning models. Findings are published in the Lancet Regional Health – Americas. The National Institutes of Health supported the research.

Up until now, no other study has looked at harnessing the power of machine learning models to help predict dementia risk among the historically understudied American Indian/Alaska Native population, as defined by the U.S. Census Bureau.

Machine learning models, which enable computers to make predictions or decisions using vast datasets without explicit programming for each task, enhance efficiency, accuracy and scalability in analyzing large datasets.

The population of older American Indian and Alaska Native adults is projected to increase nearly three-fold between 2020 and 2060. With dementia being a leading cause of disability and mortality in this age group, this debilitating condition is an increasing concern in this community.

In addition to numerous ailments like cognitive decline, weakened immune system and depression, dementia has far-reaching societal impacts. It takes a toll on family members emotionally, incurs substantial medical expenses and contributes to a general decline in quality of life.

Jiang and colleagues took seven years of data from the Indian Health Service's National Data Warehouse and related electronic health records databases and divided the data into a five-year baseline period (2007 to 2011) and a two-year dementia prediction period (2012 to 2013). The study included nearly 17,400 American Indian/Alaska Native adults aged 65 years or older who were dementia-free at the baseline, of whom almost 60 percent were female.

Over the two-year follow-up, 611 individuals (3.5 percent) were diagnosed with dementia. Four machine-learning algorithms were evaluated and compared based on their data preprocessing efforts and model performance. Of the three top-performing models the team developed, 12 of the 15 highest-ranked predictors for dementia were common across the three models. Importantly, several novel predictors of all-cause dementia, such as health service utilization, were identified across these algorithms.

Additional authors include Kayleen Ports, a former UC Irvine master's student, and Jiahui Dai, a current graduate student researcher, both from Wen Public Health; Kyle Conniff, a recent UC Irvine PhD graduate in statistics; and Maria M. Corrada, a professor of neurology in the UC Irvine School of Medicine. Spero M. Manson, a distinguished professor, and Joan O'Connell, an associate professor, with the Centers for American Indian & Alaska Native Health at the Colorado School of Public Health also contributed to the study.

The National Institutes of Health AIM-AHEAD (Artificial Intelligence/Machine Learning Consortium to Advance Health Equity and Researcher Diversity, 1OT2OD032581) and the National Institute on Aging (R01AG061189) provided funding for the study.

Awards-Winners : computer-vision-conferences.scifat.com/awards-winners

Friday, April 4, 2025

Artificial intelligence tool predicts virus outbreak hotspots

A new artificial intelligence tool could aid in limiting or even prevent pandemics by identifying animal species that may harbor and spread viruses capable of infecting humans.

Created by Washington State University researchers, the machine learning model analyzes host characteristics and virus genetics to identify potential animal reservoirs and geographic areas where new outbreaks are more likely to occur. The model focuses on orthopoxviruses - which includes the viruses that cause smallpox and mpox.

The researchers recently published a study on their work using the model in the journal Communications Biology. Their findings could help scientists anticipate emerging zoonotic threats and, importantly, be adapted for other viruses.

"Nearly three-quarters of emerging viruses that infect humans come from animals," said Stephanie Seifert, an expert in viral emergence and cross species transmission and an assistant professor in the WSU College of Veterinary Medicine's Paul G. Allen School for Global Health who helped to lead the project. "If we can better predict which species pose the greatest risk, we can take proactive measures to prevent pandemics."

The model identified Southeast Asia, equatorial Africa, and the Amazon as potential hotspots for orthopoxvirus outbreaks. These regions not only have high concentrations of potential hosts but also overlap with areas where smallpox vaccination rates are low. While the smallpox vaccine provides cross-protection against other orthopoxviruses, vaccination efforts stopped after smallpox was eradicated in 1980.

The study also identified several animal families as likely hosts for mpox, including rodents, cats, canids (dogs and related species), skunks, mustelids (weasels and otters) and raccoons. The model correctly excluded rats, which have been shown in laboratory studies to be resistant to mpox infection.

Katie Tseng, a veterinary medicine graduate student and the study's first author, noted the model not only demonstrated higher predictive accuracy than previous models, but it can be useful in predicting hosts for other viruses as well.

Pilar Fernandez, a disease ecologist and assistant professor in the Allen School who helped to lead the project with Seifert, said previous machine learning models used to predict potential hosts for orthopoxviruses relied on the ecological traits of animals, such as habitat and diet, and other characteristics that influence their interactions with the environment, such as resource use and survival. While effective, these models ignored a crucial part of the equation – the genetic makeup of the viruses.

"Previous models were more based on the characteristics of the host, but we wanted to add the other side of the story, the characteristics of the viruses," Fernandez said. "Our model improves the accuracy of host predictions and provides a clearer picture of how viruses may spread across species."

Orthopoxviruses typically cause small, localized outbreaks, but recent events, including the global spread of mpox in 2022, have raised concerns about these viruses establishing new endemic areas and spreading through new animal reservoirs.

Identifying possible reservoirs is key to anticipating spillover events, however, accomplishing that through traditional field sampling is a resource-intensive and impractical endeavor. The new model simplifies that task and can be used to target wildlife surveillance efforts.

"If you are looking for the reservoir for mpox virus in Central Africa, that's one of the most biodiverse places on Earth, so where do you start?" Seifert said. "If we can use these machine learning models to help us prioritize sampling efforts, then that's going to be really beneficial in identifying where these viruses are coming from and in understanding the risks they pose."

The research team also included Heather Koehler, an assistant professor in the School of Molecular Biosciences who has extensively studied mpox. Daniel J. Becker, University of Oklahoma; Rory Gibb, University College London; and Collin Carlson, Yale University, also contributed as members of the Viral Emergence Research Institute, a collaborative network of scientists studying host-virus interactions to predict virus spread on a global scale that is funded by the National Science Foundation. The group includes experts in data science, computational biology, virology, ecology, and evolutionary biology.

Awards-Winners : computer-vision-conferences.scifat.com/awards-winners

Tuesday, April 1, 2025

New algorithm unlocks the power of quantum machine learning

Researchers have developed an algorithm that modifies classical machine learning techniques for use on quantum computers. Their approach enables training on quantum data rather than conventional data encoded as sequences of 0s and 1s. The team tested their method on simplified tasks and found that it performed as expected, paving the way for significant advancements in quantum-enhanced machine learning.

“Quantum machine learning is a field that exploits the power of quantum computing to enhance machine learning techniques and improve computational efficiency,” Yudai Suzuki, one of the authors of the study, said in an email. “By leveraging quantum-mechanical properties such as entanglement, quantum machine learning has shown a potential to outperform conventional machine learning models.”

Quantum computing meets machine learning

Machine learning is a powerful tool that allows computers to analyze input data, recognize patterns, and make predictions without explicit programming. It has found applications across many fields, from facial recognition and natural language processing to drug discovery and material science.

Traditional machine learning algorithms work by identifying relevant features in data, progressively improving accuracy through repeated training cycles. While these techniques have revolutionized computing, they have so far been confined to classical computers, which process information sequentially using bits that represent either 0 or 1.

Quantum computers, by contrast, operate on fundamentally different principles. They use quantum bits — qubits — which, unlike classical bits, can exist in superpositions of 0 and 1 simultaneously. This allows them to perform multiple calculations in parallel, theoretically offering an enormous speed advantage for certain types of computations. Additionally, qubits can be entangled, meaning the state of one qubit is directly correlated with another, regardless of distance. These unique properties open the door to a new kind of machine learning that could surpass classical techniques.

Moreover, quantum computers can work directly with quantum states — highly abstract mathematical representations that contain complete information about a quantum system. This capability is particularly useful for tasks such as simulating quantum phenomena, where encoding data as quantum states provides a more natural and efficient approach.

Since quantum computers process information differently, classical algorithms must be adapted to function efficiently in a quantum environment. The new study, published in Advanced Quantum Technologies, addresses this challenge.

More specifically, their approach targets feature selection, a key step in machine learning where the algorithm determines which parts of the input data are most relevant to making accurate predictions. In the context of quantum computing, this means identifying meaningful information within a quantum state.

“To ensure effective performance in machine learning tasks in general, identifying meaningful and informative features is crucial,” explained Suzuki. “This principle also applies to quantum machine learning, and several proposals have explored feature selection in this context. However, existing methods are limited to classical inputs, whereas quantum machine learning can also process quantum data. To bridge this gap, we propose a new feature selection scheme that applies to both quantum and classical data.”

By integrating quantum mechanics with principles from traditional computer science, the researchers developed a strategy that maximizes efficiency at each stage of machine learning, ensuring that only the most relevant features are selected.

“One of the most significant findings is that our scheme can find relevant and important features even for quantum data tasks,” Suzuki said. “We numerically validated its effectiveness on some tasks. To the best of our knowledge, this is the first work to propose a feature selection scheme applicable to quantum data. These results indicate the potential that such quantum machine learning-oriented feature selection could be practically useful.”

Overcoming challenges and looking ahead

While the new algorithm marks a significant step forward, implementing quantum machine learning in real-world applications still faces practical challenges. One limitation is the computational cost associated with post-processing quantum data using classical resources, which could grow rapidly as system size increases. Another concern is the impact of noise in quantum hardware, which can introduce errors in calculations.

So far, the team has tested their method on simplified problems that mimic aspects of real-world quantum systems but are significantly less complex. The next step will be to apply the algorithm to more realistic scenarios, including analyzing experimental quantum data and testing it on larger quantum devices.

Awards-Winners : computer-vision-conferences.scifat.com/awards-winners

Saturday, March 29, 2025

Army eyes artificial intelligence to enhance future Golden Dome

The U.S. Army is looking to increase autonomy through artificial intelligence solutions to reduce the manpower needed to manage Golden Dome, President Donald Trump’s desired homeland missile defense architecture, the service’s program executive officer for missiles and space said this week.

As the Army contributes a large portion of the in-development air and missile defense architecture for Guam, it is looking to adapt those capabilities for a Golden Dome application, Maj. Gen. Frank Lozano told Defense News in an interview at Redstone Arsenal on Wednesday amid the Association of the U.S. Army’s Global Force Symposium in Huntsville, Alabama.

Some of the Army’s major contributions to the Guam Defense System include new modernized radars, an emerging Indirect Fire Protection Capability and its new Integrated Battle Command System, or IBCS.

“What we’re trying to do is three things,” Lozano said. “We’re wanting to integrate more AI-enabled fire control so that will help us reduce the manpower footprint. We’re wanting to create more remotely operated systems so that we don’t have to have so many operators and maintainers associated with every single piece of equipment that’s out there.”

And, he said, “We need to have more autonomously operated systems.”

Currently, the Army typically has a launcher with a missile and a launcher crew consisting of at least two to three soldiers.

“In the Golden Dome application, we would likely either have containerized missiles — think box of rockets — or we might actually put rockets and missiles in the ground,” Lozano said. Those systems would require less frequent upkeep, as a smaller manpower footprint means status checks might only happen every couple of weeks, and test checks would be conducted remotely, he said.

In order to work on such capability, the Army is planning to use what it learns from maturing the Guam Defense System, which will become operational in roughly 2027 with Army assets. The service will also pivot its Integrated Fires Test Campaign, or IFTC, from a focus on testing the Guam architecture incrementally to how to inject autonomy and AI into those systems for Golden Dome beyond 2026.

The IFTC in 2026 is considered the Guam Defense System “Super Bowl,” Lozano said. Then, beyond 2027, he said, “If we’re called upon to support Golden Dome initiatives, we need to have those advanced AI, remotely operated, autonomous-based formations and systems ready to go.”

To begin, the Army will be focused on defining the functions that human operators perform at all the operator terminals within an IBCS-integrated fire control center or at a particular launching station, Lozano said.

Once those functions are defined, Lozano said, the Army will have to define the data sources that drive action.

“We have to create the decision rubric that assesses and analyzes that data that then drives a human decision, and then we have to code AI algorithms to be able to process that information and make the right decision,” Lozano said. “There will be trigger points where the software has to say, ‘I’m not authorized to make that level of decision. It’s got to go back to the human and deliver.’”

For the first time, the Army’s Program Executive Office Missiles and Space is interacting with many new market entrants in the AI realm to work on the effort.

Awards-Winners : computer-vision-conferences.scifat.com/awards-winners

Friday, March 28, 2025

Data and artificial intelligence: the fuel behind space discovery

As maintaining leadership in space is a primary goal, particularly across the United States Space Force, NASA and other federal agencies, the U.S. remains focused on space exploration as a critical domain for missions, science investigation and national security.

One way to sustain a leading position is by achieving data dominance, leveraging artificial intelligence (AI) tools such as machine learning (ML) algorithms onboard space missions to facilitate and enable real-time decision-making.

These technologies can be used for engineering analysis and opportunistic science measurements. Data analytics and ML algorithms can also optimize resources, prioritize data to send back to Earth and identify patterns promptly.

The goal of these strategies is to develop spacecraft capable of real-time situational analysis, enabling them to make autonomous decisions and further optimize space missions. Developing and achieving autonomous science and exploration spacecraft requires a fundamental shift in the approach to space exploration. And beyond that, space organizations must navigate several technical considerations to successfully implement this vision, including environmental constraints and adapting solutions for specific mission objectives.

Data-fueled space missions

Data analytics and ML algorithms are a driving force behind space missions. They can optimize resources, such as fuel and energy usage, assist in planning and scheduling processes of observation strategies for in-space telescopes and support the prioritization of the data to first send back to Earth.

While ML algorithms on Earth can help identify patterns or correlations in massive datasets promptly, (Earth science mission teams are not scaling with the large amounts of data on hand) ML models onboard a spacecraft could help make missions even more efficient. A ML-enabled spacecraft on a life detection mission, for example, could analyze the data it gathers, identify the organic compound signatures in real-time and in the end prioritize other sampling locations without ground-in-the-loop intervention.

A long-term goal of this approach would be to have in situ analysis with spacecraft operating and analyzing in real-time, making autonomous decisions that prioritize scientific goals without depending solely on Earth-based operations.

Imagine a spacecraft on Saturn’s moon Enceladus, collecting data from the plumes being ejected at the south pole, then analyzing the data onboard and reprioritizing other operations without having to wait for a transmission from scientists on Earth — this could all be based on data collections using onboard AI-based models, software analysis and edge computing.

While this onboard implementation would help make decisions in situ to optimize resources and scientific returns, several hurdles must be overcome to see this vision through for a more efficient future.

Space exploration: data and challenges

One primary challenge in implementing AI-enhanced spacecraft is the limited onboard computing power, constrained by strict power and weight limitations that make distributing power among communication, mobility, running onboard experiments, computing and much more a difficult balancing act. Also, the “space-proofing” process — including thermal control, radiation shielding, and protection from meteoric and orbital debris complicates hardware development — and raises costs.

Bandwidth limitations and communication delays present another challenge in data transmission. Moreover, when the planetary target is not in Earth’s direct line of sight, communication becomes entirely impossible with traditional spacecraft.

Of all things, trusting the AI-driven strategy is a major challenge, particularly for life-detection missions. ML models are often seen as “black boxes,” making it difficult for scientists to fully trust the appropriate algorithms’ outcomes.

Embracing AI and ML in space exploration inspires true optimism and curiosity, and scientific discovery. To achieve this future, the industry must prioritize solutions like the development of hardware that allows real-time AI computations, advancement of data transmission tools and continuing investments in the Deep Space Network (DSN) to further enhance the efficiency of data transmission for missions.

One of the main difficulties is that the space industry must show that new hardware is truly impactful. Space missions rely on flight heritage — to prove that this new hardware works, industry needs a process to test and demonstrate tech that will have the new hardware onboard and show the success of the mission.

The testing of data-driven-enabled data prioritization must occur too — currently, space missions are designed to collect the amount of data that can be sent back to Earth. With AI-enhanced spacecraft, a fundamental shift can occur, as the ability to transmit data back to Earth will no longer be the significant bottleneck as data prioritization could be implemented.

The end goal is to collect as much data as the instrument onboard can, then have a smart algorithm onboard to prioritize the “most interesting” data to send back to Earth. More opportunities to test algorithms onboard during simulation and on low-risk science missions will bolster solutions’ technology readiness level.

Space missions have relied on pre-programmed instructions and extensive ground-in-the-loop analysis. This approach is a total change to the paradigm of space research and development, but becoming data-driven in space is necessary for success, especially when exploring targets farther away in our solar system.

Data-driven future

Private sector collaboration is key to helping transform space missions — involving expert perspectives would provide innovative solutions and strategies to help space teams develop spacecraft that can process, transport and interpret critical data.

This collaboration could be leveraged to enable real-time AI computations directly onboard the spacecraft, while also enhancing data processing pipelines for operations teams, from data collection to prioritization. Moreover, this collaboration could help accelerate the development of AI processors for space applications, ensuring they remain radiation-proof and extremely power-efficient. These collaborations are already occurring, such as NASA and IBM’s AI partnership.

Space agencies and the overall space industry must also implement an intelligent data collection solution, or data processing pipeline, that includes data collection, data labeling, analysis and then managing it appropriately so teams can access and make decisions in near real-time for mission-critical operations.

Data can also be leveraged in ML models for various applications, including anomaly detection, hardware failure prediction, and science data analysis. This can be done through training models on Earth and then fine-tuning them for specific space mission targets.

Using big data more effectively will also allow teams on Earth to develop visualization and simulation tools. These could involve digital twin investigations — virtual replicas of spacecraft and planetary environments to simulate missions and test algorithms before deployment — leading to smarter, more decisive actions for mission operations.

Cultivating a data-driven environment is not just about implementing next-gen tools, but it is a catalyst for the next frontier of space discovery and exploration.

Advancing AI-enhanced space exploration requires interdisciplinary collaboration (among AI experts, software engineers, astrobiologists and so forth) ensuring tools and models are adaptable and scalable. As computing power and onboard capabilities improve, data-intensive tasks (such as spectral analysis through ML) could increasingly be performed in space, to enable real-time insights and collaborative science discoveries, unlocking the next frontier of space exploration.

Awards-Winners : computer-vision-conferences.scifat.com/awards-winners

Friday, March 21, 2025

The conquest of Artificial Intelligence and its perils

It still seems incredible to us to think about how quickly artificial intelligence has advanced and the changes it has made in such a short time in humanity. It feels like it's been longer, but it was only on November 30, 2022, when the company OpenAI launched its ChatGPT project. This conversational bot changed everything from the processes of entire companies to the way education was conceived. Its conquest has been of such magnitude that the Nobel Prize committee awarded John Hopfield and Geoffrey Hinton the prize in Physics for discoveries and inventions that laid the foundations of machine learning and artificial intelligence. That is the dimension of the conquest of artificial intelligence concerning the methods that have changed, the tools available to us today, and present and future inventions.

Referring to the universe of Artificial Intelligence is like entering a deep sea full of complications that generate solutions. The award-winning researchers have used tools from physics to develop methods that are the basis of today's powerful machine learning. This gave the Nobel committee enough reasons to choose them. It is not a minor issue that these have been the merits to win such a prestigious award.

Professor Hopfield conducts his research at Princeton University, while Professor Hinton works at the University of Toronto. Both laureates applied fundamental concepts from statistical physics to designing artificial neural networks that function as associative memories and find patterns in large batches of data.

We may believe that Artificial Intelligence processes are complex — and they undoubtedly are — but the ease of using them in our daily activities is what allowed their applicability to be so popular and become so widespread. That is the paradigm through which Artificial Intelligence works, which is fast and accurate. For example, facial recognition by our mobile phones is just one example of this applicability. Our biometric data is used today to recognize us when entering and exiting an office or when accessing our bank accounts or various digital platforms. Something that we used to imagine as a science fiction fantasy, today is part of our daily lives.

The achievements of Artificial Intelligence will be as relevant as the progress generated by the Industrial Revolution. The leap is as great as the one experienced when artisanal production was exchanged for factory production. And, although I am sure that there was resistance and voices that criticized or went against it, in the face of progress there is no alternative. There are those who embrace it and those who do not, those who jump on the trend and those who are left behind.

Of course, the benefits and their effects are already evident. Professor Hinton predicted that artificial intelligence would "end up having a huge influence on civilization and lead to improvements in productivity and healthcare." This gives hope to people who suffer or see suffering from diseases that today seem to have no cure.

However, we cannot be naïve. In addition to applauding the progress and advances that Artificial Intelligence brings, we must also open our eyes, pay attention, and realize that not everything is pristine and perfect. It is necessary to address the concerns that are expressed about a series of possible bad consequences, such as the alteration of photographs — we already saw the Father in a very elegant white coat from Dolce & Gabbana — as well as other types of fraud and bad practices. We must consider the threat of these things getting out of control.

Of course, all that glitters is not gold. Although the committee awarded them the Nobel Prize in Physics, it also recognized that the science behind machine learning and artificial intelligence has its negative aspects. We need to understand that while machine learning has enormous benefits, its rapid development has also raised fears about our future. Fears that are legitimate.

Humanity in general, and each of us in particular, has a responsibility to use this new technology in a safe and ethical way for the greater benefit of human beings. We cannot cover the sun with one finger. Even Professor Hinton himself shares those concerns. He explains that he left a position at Google so he could speak more freely about the dangers of the technology he helped create. We must listen to the voices that warn about the excesses and misuses of technological advances.

Professor Hinton acknowledged being shocked by the recognition. "I'm dumbfounded. I had no idea this was going to happen," he said when contacted by the Nobel committee by phone. It is time to understand that machines learn and take charge of laying the foundations for the development of Artificial Intelligence. We thought that learning was an attribute of animal brains and now we see that artificial brains can also learn.

The work of both scientists has helped computers to be able to imitate human functions such as memory and learning. Hopfield created an associative memory in 1982, which could store and reconstruct images and other types of patterns in data. Hinton, for his part, developed a method that allows a machine to find properties in data autonomously and thus perform tasks such as identifying specific elements in images. These investigations and achievements paved the way for artificial intelligence systems such as ChatGPT.

"We have no experience of what it's like to have smarter things than us," he said. "It's going to be wonderful in many ways, but we also have to worry about a number of potential negative consequences." And, yes, the professor himself declares: "My guess is that, within five or 20 years, there will be a 50% chance that we will have to face the problem of artificial intelligence trying to take control of our lives."

To do this, we need to get down to work today and not 50 years from now. We need to lay the foundations for a responsible and ethical use of Artificial Intelligence, to enjoy its benefits, and to restrict its dark aspects. That way, we can enjoy the conquests of Artificial Intelligence before allowing it to conquer us.

Awards-Winners : computer-vision-conferences.scifat.com/awards-winners

Wednesday, March 19, 2025

Taco Bell Parent Accelerates AI Innovation With Nvidia

Quick-service restaurant (QSR) giant Yum! Brands has partnered with Nvidia to develop and scale artificial intelligence (AI) technologies for its restaurants.

These technologies will be deployed at the QSR company’s KFC, Taco Bell, Pizza Hut and Habit Burger & Grill restaurants, the companies said in a Tuesday (March 18) press release.

The partnership will help scale Yum! Brands’ existing proprietary AI-driven restaurant technology platform Byte by Yum!, according to the release.

“This partnership will enable us to harness the rich consumer and operational datasets on our Byte by Yum! integrated platform to build smarter AI engines that will create easier experiences for our customers and team members,” Joe Park, chief digital and technology officer at Yum! Brands and president of Byte by Yum!, said in the release.

The AI solutions will include voice-automated order-taking AI agents for drive-thru and call center operations, computer vision for optimizing drive-thru efficiency and back-of-house labor management, and AI-driven analytics and agents for assessing restaurant performance and generating personalized action plans for restaurant managers, according to the release.

Yum! Brands has already piloted several AI solutions in select Taco Bell and Pizza Hut locations and, after the success of the pilot, plans to roll out the technology to 500 Pizza Hut, Taco Bell, KFC and Habit Burger restaurants during the second quarter, the release said.

Looking ahead, the company plans to continue expanding its use of AI and integrate more advanced AI models, developing solutions that will be built with the latest Nvidia software and be proprietary to Yum!, per the release.

Andrew Sun, global director of retail, CPG and QSR business development at Nvidia, said in the release that working with the Yum! Brands team and platform to integrate Nvidia AI software “breaks barriers to AI innovation in the restaurant industry — delivering real-time, context-aware intelligence, powered by a scalable inference platform.”

Yum! Brands said in February that the Byte by Yum! AI-driven platform was already in use at 25,000 international locations and will be rolled out throughout its global locations.

In November, the company said that it had processed over 2 million successful orders with the drive-thru voice AI system it had in place in over 300 Taco Bell stores in the U.S. and that many franchisees were eager to test this innovation at their own locations.

Yum! Brands also said in November that early, limited pilots of AI-powered marketing campaigns had delivered double-digit increases in customer engagement compared to traditional digital marketing campaigns.

Awards-Winners : computer-vision-conferences.scifat.com/awards-winners

Tuesday, March 18, 2025

AI Investments Expected to Shift to Inference While Growing Faster Than Forecast

The impact of reasoning AI models from DeepSeek and OpenAI is reportedly expected to shift the focus of artificial intelligence (AI) investments while also boosting AI spending overall.

While the debut of the DeepSeek models led observers to question the need for investment in AI infrastructure, it also led to a greater focus on reasoning models, which require greater spending on inference, Bloomberg reported Monday (March 17).

As a result, Bloomberg Intelligence now expects the investments in AI by hyperscale companies like Amazon, Meta and Microsoft to increase faster than it previously forecast, with more on that money being spent on running AI systems after they have been trained, rather than on data centers and chips, according to the report.

These companies are expected to spend $371 billion on data centers and computing resources in 2025 — 44% more than they spent in 2024 — and $525 billion a year by 2032, the report said.

By 2032, nearly half of all AI spending will be directed toward inference, as reasoning models enable companies to make more money from software, per the report. At the same time, the share of investment directed toward training is expected to drop from 40% to 14%.

DeepSeek shook up the AI world in late January when it released AI models that performed on par with OpenAI’s and Google’s top models but at a fraction of the cost and with far fewer of Nvidia’s GPUs.

Shortly after the release of the DeepSeek AI model that rocked the AI world, Meta CEO Mark Zuckerberg said during an earnings call that the U.S. AI industry is shifting toward AI processing, or inference, as reasoning AI models rise in popularity.

OpenAI released in February what it called its “most cost-efficient” reasoning AI model, the o3-mini, saying it is a “small” but “powerful and fast” model that outperforms earlier models especially in science, coding and math, and comes in three reasoning levels: low, medium and high for tougher tasks.

The model is part of OpenAI’s o1 series, which can reason through tasks but takes longer to respond than non-reasoning models. Reasoning models can also tackle tougher tasks and solve harder problems.

Awards-Winners : computer-vision-conferences.scifat.com/awards-winners