AWS report: Generative AI overtakes security in global tech budgets for 2025

Generative AI instruments have exceeded cyber as the highest budget priority for global IT leaders in 2025 Comprehensive new research It was released today Amazon Web Services.

This AWS Generative AI Adoption IndexOrganizations that surveyed by 3,739 high decisions in nine countries, 45% plan to make a significant change in corporate technology strategies to capitalize traditional IT investments (30%) such as security tools.

“I don’t think this is caused by concern.” “I interpret the security of customers as a mass priority. The AI ​​budget priority is that it is so large that customers see so many uses for the AI, which manages the special result.”

An extensive survey conducted between the United States, Brazil, Canada, France, Germany, India, South Korea and the United Kingdom, these technologies show that 90% of these technologies have reached 90% of the total infections. More delicate, 44% have already been transferred to the production of production outside the experimental phase.

60% of companies have already appointed Chief AI employees as C-Suite has already become a C-Suite

Among organizations, such as AI initiatives, new leadership structures are created to manage complexity. In the report, 60% of organizations have already appointed a special EU executive as the EU President (CAIO), which is another 26% to do this in 2026.

This executive level reflects the recognition of the strategic importance of the EU, although the research says that this will not yet have official AI transformation strategies, and offering potential problems in the management of quarters.

“A thoughtful change management strategy will be critical,” he said. “The ideal strategy must apply for operational model changes, data management experience, talent pipelines and scaling strategies.”

Companies Average 45 AI experiments will only reach 20 users in 2025: the problem of production gap

The organizations were expected to have an average of 45 AI experiences in 2024, but by 2025 to end users by 2025 to end users by 2025.

“I think it is very fast and higher than 40% to see more than 40% to go to production for something relatively new,” said Pathak. “He said that it was completely used by customers to produce a scale, and I think it continues to accelerate.”

As a basic obstacle to the production of talent, talent deficiency, 55% of respondents are not the biggest problem of a capable generative AI workforce.

“To successfully unlock to obtain production to obtain a success, they really understand how the work goals they are trying to drive, then how the EU will interact.” “Really, when you are combined with your customers with AI, which you can manage a business outcome associated with your business and customers.”

92% of organizations will hire a AI talent in 2025, while applying training for 75% bridge skills

In order to touch the skills, organizations are watching dual strategies for domestic training and foreign recruitment. Inquiry, 56% of organizations have already developed generative AI training plans, and the other plans to do so by the end of 2025.

“For me, this is the mind for customers,” said Pathak said about the lack of talent. “We are confident that we bring our teams and our employees together and take the opportunity to a maximum increase.”

He noted the ability to adapt, not specific technical skills.

Talent boosts are 92% of the exercises for aggressive hiring, 92% of organizations for roles that require generative AI expertise in 2025. In the quarter of the organizations, at least 50% of these skills will require.

Financial Services Hybrid AI Revolution: only 25% of companies Solutions from scratch

The existing AI solutions or models that are used to create a long time to create ownership AI solutions or levers of models seem to be resolved in favor of a hybrid approach. Only 25% of organizations plans to plans to resist at home from scratch, 58% will develop applications in special applications and 55% in pre-existing models.

It traditionally represents a remarkable change for industrial areas known for special development. The report plans to use the historical advantages of financial services, 44% of companies plan to use the box solutions – their historical advantages.

“A large number of customers still build their models,” said Pathak. “It is said that there are many abilities and investments that have excellent basic models with perfect starting points and worked hard to make sure of customer information.

Added that companies still can still use property science when using existing basic models: “Customers understand the world with items such as dwarfs can achieve the benefits of the world’s property [Retrieval-Augmented Generation] And with customization and delicate adjustment and model distillation. “

India received Western markets in 64% of Global AI with South Korea after 54%

Although a global trend in the generative AI investment, the work has revealed regional change in the adoption. The United States, globally, 45%, India (64%) and South Korea (54%), (64%) (54%) (54%) aligned generative Ai investments showed 44% of the prioritative EU investments.

“We see our mass adoption in the world,” he said. “I thought it was interesting to have a relatively high consistency on the global side.

65% of organizations will trust third party vendors to accelerate the implementation of EU in 2025

When organizations manage a comprehensive AI view, organizations rely on foreign experience. The report plans to adopt 15% and 50% to accept 65% of organizations in 2025, 65% in 2025 to adopt a mixed approach to vendors and local teams and foreign partners.

“It’s a very ‘and’ type of relationship for us,” said Pathak AWS’s approach to support both special and pre-established solutions. “We want to get acquainted with the clients where customers are. A model provider has a great partner ecosystem of service providers and system integrators.”

Now that is important to move or stay behind

For those who hesitate to accept the generative EU, Pathak, a hard warning: “I really think they are going to be based on customers or the risk of peers behind.”

The area of ​​innovation accelerated in the area: “The rate and the level of improvement of AI technology and the reduction of results and the results of the results will be impossible, it will be possible to seem like old news in almost three to six months.”

This thinking is widespread among the sectors. “We see such a fast, such a massive width,” said Pathak. “We see adjustable industry, financial services, health, governments, large enterprises, beginnings. The current product of beginners is almost only managed by the EU.”

First approach to AI success

This AWS report Generative increases the portrait of the fastest evolutionary to the main business infrastructure. As organizations have reached a destination to ensure budget priorities, reconstructed leadership groups and AI talent, as a decisive hole in the company’s adoption of the enterprise.

Technological gold hurts, the most successful applications are probably more than organizations that are more than technological innovation, the outcome of a cruel focus. As Pathak stressed, “AI is a powerful tool, but you should start with your work goal. What do you try to start as an organization?”

In the end, developing companies are definitely the largest AI budgets or the most advanced models, but most effectively use the AI ​​to solve real business problems with unique information assets. In this new competitive landscape, the question is whether the organization can change the AI, but how many organizations can change the transformance of organizations as soon as possible.