The phase of accelerated transformation in which the technology sector finds itself is forcing businesses and consumers into a rapid adaptive response. According to sector studies, the year 2025 is presented as a “starting point” in this process that will end by change the configuration of digital infrastructures that support the daily activities of companies and their clients, in addition to the way they interact with new technologies.
In this framework, a survey carried out by the research department of the technology firm Akamai detected the six technological trends that will impact both users and organizations in 2025.
Among the main elements predicted are Artificial Intelligence (AI), cloud infrastructure, cybersecurity, language models and post-quantum cryptography.
But this does not mean that all trends have a positive biassince some have to do with the generation of higher risks for all types of users due to the greater sophistication that cybercrime is achieving.
For example, they anticipate that next year deepfakes, targeted scams and social engineering will become widespreadwhich will become increasingly accessible even to common cybercriminals.
And since the existing cloud infrastructure is under pressurethey also predict that 2025 will be the year in which The cracks that are currently visible in that system will become failures.
What are the six technological trends for next year?
-AI agents redefine the Web: AI agents will transform the way we interact with the web, allowing tasks such as scheduling appointments, purchasing products or paying bills to be done automatically, reducing our dependence on screens. Chatbots will evolve to perform low-risk tasks more efficiently. For example, your healthcare provider’s chatbot that today can tell you how to navigate screens and menus to schedule an appointment will soon do it for you, offering you some available appointment times.
-Social engineering cyber threats are becoming widespread: By 2025, cyber threats such as deepfakes, targeted scams and social engineering will become widespread, accessible even to common cybercriminals. The tools to create convincing deepfakes in all of these interactions are already, or will soon be, available to criminals who will use them to scam you or reveal valuable information.
-Existing cloud infrastructure is under pressure and 2025 will be the year cracks become failures: The growing computing demands of technologies such as spatial computing, AI agents and smart urban infrastructure will require a significant overhaul of our existing cloud infrastructure. The current centralized model, designed for the early 2000s Internet, will not be able to support the scale, latency requirements, and data demands of these new technologies. To accommodate the rapid growth and evolution of the digital landscape, we will need to invest in highly distributed cloud infrastructure. This will involve deploying powerful computing points closer to users and devices, enabling real-time interactions, low-latency applications, and the ability to handle massive amounts of data. If we do not adapt our infrastructure, we will limit the potential of these emerging technologies and hinder innovation in various industries.
-2025 will be the year in which post-quantum cryptography becomes widespread: Post-quantum cryptography (PQC) guarantees security against possible attacks by quantum computers. This cryptography protects information from vulnerabilities that could arise with the advancement of quantum computing, which is capable of performing calculations at speeds much higher than those of ordinary computers. In 2024 we receive the long-awaited post-quantum encryption standards and 2025 will be the year of quantum readiness for most CIOs and CISOs. Likewise, organizations must begin preparing for Q-Day.
-LLM security risks become tangible: Linguistic models (LLM) are artificial intelligence systems that understand and generate human language in a coherent way. While we have already seen public disclosures of some of the exploitable flaws in LLM, the rate and severity of these will begin to increase. LLM represents a huge attack surface that malicious actors can exploit, so companies will begin to question their LLM plans.
-Small language models steal the scene: Small language models (SLM) are more compact versions of artificial intelligence created to process and generate natural language efficiently, using few computational resources. These provide personalized information and carry out specific tasks without requiring complex infrastructure. Small Language Models (SLM) are poised to become extremely popular within enterprises in 2025. Their ability to provide personalized information, reduce reliance on top-of-the-line GPUs, and optimize performance will make them an attractive option for enterprises seeking They seek to improve their products and services. Additionally, the growing emphasis on data privacy will drive enterprises to adopt SLMs that are easier to implement on-premises and easier to secure. Furthermore, the modular design and scalability of SLMs will allow companies to adapt them to their specific needs and changing requirements.
Source: Ambito
David William is a talented author who has made a name for himself in the world of writing. He is a professional author who writes on a wide range of topics, from general interest to opinion news. David is currently working as a writer at 24 hours worlds where he brings his unique perspective and in-depth research to his articles, making them both informative and engaging.
