Agentic AI vs generative AI: why the futures not just smarter its bolder

You give it a direction—“improve customer churn”—and it starts to act. It looks at retention data, cross-checks CRM logs, generates hypotheses, triggers outreach campaigns, and, crucially, updates its approach as new data rolls in. Agentic AI uses reasoning, decision-making algorithms, and environment-based data to act and adapt.

African Development Bank approves financing to advance Rwanda’s universal energy access

However, public research on audio recognition and emotional audio generation remains limited. Get insights and exclusive content from the world of business and finance that you can trust, delivered to your inbox. The future of AI factories isn’t just technical—it’s democratized, collaborative and fundamentally human-centric in its design, ensuring that anyone with domain expertise can contribute to the agent economy regardless of their coding background. It’s this final capability—turning understanding into action—that makes agents the highest-order output of AI factories. When McKinsey projects that AI could add $4.4 trillion annually to the global economy, they’re not referring to passive intelligence or token production alone, but to the automated execution that agents enable across industries.

This is more than mere automation — it’s intelligent, proactive management and autonomous automation of complex tasks. Beyond invoice handling, AI agents can significantly ease tasks such as account reconciliations, audit preparation, fraud detection and cash flow forecasting. By automating these critical yet repetitive processes, accountants not only save time and resources, but they also dramatically reduce the risk of human error, gain real-time visibility into financial health, and enhance their responsiveness to financial anomalies. Ultimately, this means accountants can shift their focus from managing day-to-day operations to more strategic roles, offering deeper insights and advisory services that drive greater business value.

UK could be forced to step back on data surveillance demands

Modern AI factories represent the culmination of this evolution—producing autonomous agents that convert intelligence and tokens into direct action. Unlike previous outputs that inform decisions, agents execute them, closing the loop between insight and outcome. This means implementing frameworks that monitor agent behavior, explain their decisions and maintain compliance with regulatory requirements. At any moment, this system should be able to produce reports that provide total transparency as to what their agents have done and why they took those actions. Underpinning any proper agent architecture is a comprehensive governance layer that must ensure all AI agents are closely tracked, fully auditable and completely secure. A bank could never tolerate an AI agent approving a loan for one person while declining the same loan for another person with largely the same application credentials.

• These research approaches are now out of university labs and are available in public domain for everyone to try in the form of new models.
• Imagine an operations department where AI isn’t just used in workflows but actively manages them.
• What’s needed now is global dialogue on standards, data governance, and sustainable implementation, and WHX Tech provides the ideal platform for that,” said Dr. Shah.
• If you’ve ever played around with any LLM like ChatGPT, try to ask it the same question twice and see what happens.

Regulatory compliance simply doesn’t have room for creative interpretation. This is precisely the risk holding back many AI agent deployments today. Finally, it must optimize those workflows as it moves through its processes. This means it should be able to detect when a better approach is possible on the fly and then implement that change—if and when a human approves. The same poll question found that 50% of respondents said they were researching and experimenting with the technology, while another 17% said that they had not done so, but planned to deploy the technology by the end of 2026 at the latest.

China’s flagship global infrastructure initiative is changing in the face of potent headwinds

“Technology only works when it fits into the everyday workflows of real people. Our studies show that nearly half of healthcare workers struggle to understand the tools meant to empower them. At WHX Tech, we’re championing inclusive design, digital literacy, and public-private collaboration to build trust and scale adoption. AI-powered agents are capable of taking on a diversity of roles.

• What’s missing is an intelligent orchestration layer to ensure all agents are working together, acting in the organization’s best interests instead of freelancing as they see fit.
• By updating its virtual assistant’s core natural language processing engine to the latest GPT models, Air India achieved 97% automation in handling customer queries, significantly reducing support costs and improving customer satisfaction.
• This progression isn’t about discarding earlier outputs but integrating them.
• AI agents’ use of natural-language processing also changes the equation.
• Integrated seamlessly into familiar workflows, AI agents will quietly amplify efficiency and effectiveness while minimizing complexity for users.

These agents, when applied to consumer use cases, start giving us a sense of a future where everyone can have a personal Jarvis-like agent on their phones that understands them. Want to book a trip to Hawaii, order food from your favorite restaurant, or manage personal finances? The future of you and I being able to securely manage these tasks using personalized agents is possible, but, from a technological perspective, we are still far from that future. A majority of 1,100 tech executives (82%) responding to a recent survey from consultant Capgemini indicated they intend to integrate AI-based agents across their organizations within the next three years — up from 10% with functioning agents at the current time.

Dubai Health Authority welcomes 20 global health leaders

And that’s just the tip of the iceberg when it comes to other security threats companies are dealing with in 2025. We’re going to take a look at the current security threat landscape on this episode of Today in Tech. It’s about giving them back the 40% of their day they spend nudging, chasing, checking, and… sighing. Such advanced capabilities are driving rapid growth in the AI agent market, expected to expand from $5 billion today to approximately $47 billion by 2030, according to a study by ResearchAndMarkets.com. We continue to hear about the latest and greatest model launches from usual suspects like OpenAI, Cohere, Anthropic and Mistral.

“Currently, to automate a use case, it first must be broken down into a series of rules and steps that can be codified,” the McKinsey team said. NVIDIA continues to lead the charge in AI infrastructure, with predictions indicating a shift towards quantum computing and liquid-cooled data centers. Quantum computing advancements, particularly in error correction techniques, promise to enhance computational power and efficiency, addressing instability issues that currently limit quantum hardware. Performance and cost efficiency are further amplified by NVIDIA TensorRT-LLM optimizations, now applied to popular Meta Llama models on Azure AI Foundry. These include Llama 3.3 70B, 3.1 70B, 8B, and 405B, delivering immediate throughput and latency improvements—no configuration required. Imagine a future where an AI agent not only books your next vacation but also helps provide a shopping list based on your destination, weather forecast, and the best deals from around the web.

And when you hand off autonomy, even partially, you’re entering a zone that demands trust and control. Agentic AI is no longer just a concept; it’s quietly proving its worth across industries, paving the way for a future where technology doesn’t just assist but acts. These research approaches are now out of university labs and are available in public domain for everyone to try in the form of new models.