#6: The Evolution of Agentic AI Systems [3-min read]

Exploring #FrontierAISecurity via #GenerativeAI, #Cybersecurity, #AgenticAI.

Feb 05, 2025

"If you create a system that is more intelligent than humans and has its own objectives, then we’ve lost control. The key challenge is ensuring AI systems remain beneficial, controllable, and aligned with human values."
— Stuart Russell, Professor of Computer Science, UC Berkeley

Why Agentic AI matters to everyday life.

For most people, Agentic AI might sound like a distant, abstract concept -- something that belongs in cutting-edge research labs or high-stakes business environments. But in reality, it’s about to become an integral part of daily life, influencing everything from how we work, bank, shop, get medical care, and even consume news.

Here’s why it matters to you.

1. The future of work: Will AI be your assistant or your boss?

Agentic AI isn’t just automating tasks; it’s taking initiative -- analyzing emails, making financial decisions, scheduling meetings, and even drafting reports. If you work in customer service, finance, HR, or content creation, you might soon find yourself collaborating with AI on a daily basis.

But here’s the big question: Will AI be a helpful co-worker or will it start making critical decisions without human oversight?

What happens when AI filters job applications or manages employee performance?
Will we trust AI supervisors to make fair decisions?
If AI negotiates contracts or conducts hiring, how do we ensure it’s not biased against certain people?

2. AI in healthcare: faster, smarter, but at what cost?

Agentic AI is already transforming medicine -- scanning X-rays, predicting diseases, and even recommending treatments. AI-driven diagnostics could reduce medical errors and increase access to healthcare, especially in underserved areas.

Here’s where things get tricky.

If AI misdiagnoses a patient, who’s responsible?
If AI decides who gets a hospital bed in an emergency, is it fair and ethical?
Will AI-generated treatment plans prioritize cost savings over human care?

3. AI in banking & finance: The algorithms learning your money

Agentic AI is set to automate everything from personal finance management to stock trading. Imagine a world where:

AI recommends what you should invest in, based on real-time financial patterns.
AI denies you a loan because of hidden algorithmic biases.
AI automatically moves your money without your direct approval -- based on “optimal” strategies.

Who truly benefits in this scenario -- the user or the banks using these systems?

4. AI in information, misinformation and disinformation: who controls what you see?

We already live in an era of algorithm-driven content -- where AI decides what news articles, social media posts, and search results we see. But Agentic AI takes it further by autonomously creating, curating and distributing content.

This raises an important set of questions.

Could AI reinforce political biases by only showing content it thinks you agree with?
What happens if AI-generated misinformation goes viral faster than we can fact-check it?
Will AI-driven news make it harder to distinguish real journalism from fabricated narratives?

5. AI in national security & surveillance: who has what access?

Agentic AI is already being studied for policing, border control and cybersecurity. This means AI-driven facial recognition, predictive policing, and real-time tracking of public movements could become widespread.

If AI determines criminal risk, could it wrongfully profile individuals?
If AI flags "suspicious activity," who gets to question the system’s accuracy?
Will AI-driven surveillance erode privacy and personal freedoms before we even realize it?

KT’s Final thoughts: Why should you care?

Agentic AI isn’t coming -- it’s already here. In the next few years, it will determine who gets hired, who gets medical care, who gets financial opportunities, and what information is deemed “true” online.

Done right, it can reduce bias, increase efficiency, and create better decision-making systems.
Done wrong, it could exacerbate inequality, enable mass surveillance, and automate injustice.

This is why we need governance and accountability now, before Agentic AI is deeply embedded in society. The key is not rejecting the technology, but ensuring it serves human needs -- not just corporate or government interests.

Meanwhile, semi-autonomous AI systems are actively shaping law enforcement and immigration processes. As AI agents become more advanced, the question isn’t whether they’ll be used in these sectors -- it’s how much autonomy they’ll be given.

Innovating with integrity,
@AIwithKT 🤖🧠

AIwithKT