AI Agents - is yours named 007 perhaps?

24 July 2025

Hollywood movies popularised the 007 Agent (courtesy of Ian Fleming, the author).


Technology companies are popularising the AI Agent.


If you have not heard of Artificial Intelligence Agents or the term Agentic AI, then you will do. AI Agents will become as commonplace as having an App on your phone.


Just look at an AI Agent as a ‘coworker’ or ‘support worker’ or an ‘AI Collaborator.’ AI Agents can be seen as a helping hand. An AI agent who can perform various tasks which can help humans with their daily work from crunching numbers for example, to summarising vast amounts of complex information, to dealing with supply chains or handling detailed customer queries.


Call your AI Agent what you wish, whether it is 007, 008, 009 or OBEE1, AI Agents already come in many shapes and sizes often hidden from plain sight functioning as part of your internal organisation’s processes, doing beneficial work.


These AI Agents are trained to handle complex, multi-step tasks. These AI Agents differ from Chatbots. Chatbots are more basic in usage i.e. they use AI to simulate human conversation and provide information based on predefined rules or data sets such as set scripts. Chatbots often pop up on webpages helping with basic queries. However, if you have a more complicated query, you are often transferred to a human.


An AI Agent uses large language models to perform complex interactions and multi-step tasks.


In the future you will see AI Agents and Chatbots becoming more sophisticated with a blurring of lines between the two. Currently, AI agents are more powerful than Chatbots, in that AI Agents can perform actions without direct user input with varying degrees of capability. AI agents can be segmented into six simple categories:


  1. Simple reflex, likened to a human reflex action, these AI agents perform simple actions, responding to preset rules and conditions e.g. elevator control doors open, or close, responding to button presses.
  2. Goal based, AI agents are programmed to achieve key goals or outcomes e.g. data driven insights in healthcare diagnostics or AI Agents supporting human clinical coders with clinical coding.
  3. Learning, AI learning agents will learn continuously; they take their input from various sources e.g. ecommerce platforms handling concert ticketing will take input from demand patterns and competitor information and apply dynamic pricing. So, the price of tickets can go up if demand is high.
  4. Model based reflex AI Agents have a memory i.e. they retain a model of the environment, as well as understanding the current environment, which can help predict future action e.g. used in robotics across industry with sensors to avoid obstacles and plan alternative routes.
  5. Utility based – based on clear frameworks laid out by your organisation such as time, cost, quality, or an agent could in a car look at speed versus performance. A utility agent is dynamic, constantly learning and often operating in a complex environment it will weigh up options to achieve the best outcomes within the parameters given such as driving up revenue or performance.
  6. Hierarchical agents – AI agents use a structured chain of command to carry out tasks with an AI supervisor delegating sub tasks e.g. managing workflow tasks in Human Resources such as handling data extraction or form filling.

Governance


With the introduction of any new technology, AI Agents need to be adopted with humans in mind and clear guardrails need to be in place. The right processes and procedures for checking the output of AI Agents need to ensure there is no misinformation or disinformation, and malintent by a rogue AI agent or deliberate human sabotage. In organisations the governance steps needed also include cyber security, data privacy together with legal and regulatory policies, procedures, and clear ethical guidelines for those working with AI Agents. This will ensure that your organisation mitigates risks, protects its customers, employees, partners, and key stakeholders who will interact with the AI Agents.

 

The Future with AI Agents


So, what of things to come with AI Agents? The future seems bright if put to beneficial good for organisations and personally. The use cases for AI Agents seem endless. From business usage to your personal life, AI Assistants will spring up offering a range of products and services. AI Agents will be part of everyday appliances, robotics, and software systems.


By2028Gartner predicts 33% of enterprise software applications will include Agentic AI.


Gradations of service will be offered with for example elite AI butler services, to everyday help. These AI Agents will be made available more en masse, with 24x7x365 help. Teams of AI agents will work either together or with humans to form ‘super teams.’ The AI agents will support productivity, efficiency savings and bring about new ways of working, new cultures.


AI Agents will increasingly become more specialised.


  • For industry specialisation will be particularly with industry verticals, where we are already seeing AI help across verticals, AI Agents will become highly specialised in a particular domain, such as Healthcare or Finance, or Manufacturing. These ‘vertical AI Agents’ will sit alongside Horizontal agents who help with organisation-wide tasks which could be for example human-resource or finance related, i.e. complex tasks which apply to the whole organisation and do not necessarily have to be industry-specific.
  • At home you will see AI Agents in use such as your own personal shopper who can choose and pay for what you want whether its scanning for holidays or clothes or food shop. The AI agent will perform the full purchasing cycle of looking to choosing and buying can be done for multiple purchases in seconds with delivery scheduled in your calendar as your AI Agent learns what you like or want and it could be if you build up enough trust without even asking.


These AI Agents provided they are guided and instructed well, will never tire, or get bored. They may ‘age,’ i.e. they may need software updates but that should happen automatically dependent on the underlying software provider. Many AI Agents will self-improve or employ other AI agents to provide them with the latest updates or better capabilities. These AI Agents may sit in a hierarchy so that the ‘higher order’ AI Agents receive their crucial updates first, so they are renewed and energised to just do the right things for organisations for individuals.

In Summary



AI Agents or Agentic AI is the next level of Artificial intelligence. At the Human Digital Collaborative, we believe it can bring a wealth of opportunities to your organisation, if applied in the right area with the right guard rails in place.


We can collaborate with you to determine your AI needs using our 360° AI Service. We work in collaboration with World Class partners to create valued sustainable AI solutions which you can trust. We can even help you name some of your Agents to join your team!


Do contact us @TheHumanDigitalCollaborative.com


10 June 2025
Have you checked your Cyber security resilience capabilities?
7 February 2025
Beware the eyes of cyber - do you know who's looking?
10 January 2025
Figuring out your key forces and types of Change 
14 November 2024
Seeing & controlling the ‘C’s’ of Change
26 September 2024
Transformation twists and turns? For anyone who has undertaken a transformation programme particularly a significant, complex, multi-faceted transformation you will know that there are many twists and turns to ensuring success, particularly if the journey is taken over a number of years. Key to good transformation planning is making sure you have the support of your stakeholders, the right people in place, and the right resources to make it happen whether that be for example budget or systems, they all need to work in harmony to keep the transformation alive and help pave your way to success. Holding regular reviews is also important i.e. looking back and looking forward, looking inward, as well as outward to make sure you are still on the right path, and you keep pace with the times. A good illustration of why transformation is important is by looking at a case of a well-known company who enjoyed many years of success, but then failed to transform quick enough with challenging consequences: Kodak. Eastman Kodak (commonly know as Kodak) was incorporated as a business in 1832, and it grew to be one of the leading manufacturers of film and cameras in the world. However, it failed to recognise the significant threat of other competitors particularly in its market’s-movement to digital. This is despite one of its own employees in 1975 developing the world’s first hand-held digital camera, and another employee four years later predicting that a shift to digital photography was inevitable. Kodak executives seemingly ignored the warning signs, the advice given, and sadly it started its transformation journey too late, as over a century later since it began, this once successful company had to file in 2012 for Chapter 11 bankruptcy protection [1] in the United States. A massive decline for a brand and a business who at one time seemed untouchable in its market. A huge price to pay for taking time to transform. The story for Kodak however, is not all bleak, as Kodak can be applauded for its tenacity to stay in business. By 2016 it once again started to generate a profit by transforming and restructuring its business. There are many other business illustrations of failing to transform that I could cite, however Kodak is a memorable example of a business whose longevity unfortunately did not give it licence to move to transform at a slower pace. A number of key questions arise from this case study and perhaps take stock of your own organisation’s transformation journey: Would Kodak be more of a dominant brand for consumers nowadays if it had transformed its business earlier? If it had shifted its consumer film products and services earlier to digital? (it still has a film business with movies). Or if it had looked inward, recognising talent, and listening to employees in its own organisation? Or if it had looked outward and reacted faster to the competitor threats from Fujifilm or indeed the smartphone and device manufacturers? How the picture (pardon the pun!) may have been different today if such a dominant company with a very strong brand, had fine-tuned the twists and turns of transformation to its advantage? Transformation timing? We hear often from businesses – now is not a good time to transform, or we tried that, and it didn’t work, or those types of change (e.g. front or back- office administration process automation) works in that business sector but wouldn’t work in ours? Ask yourself are these just excuses for not taking bold steps to transform or are there genuine reasons for standing still in your organisation? The key question is can you really afford for your organisation to stand still? As a former CEO we know always said after implementing a successful transformation, ‘there’s more to do.’ In other words, his success was not going to be pinned to the mast of completing the last journey to improvement, but rather looking across the organisation to see how things could be improved further for its customers, its staff and its stakeholders.
30 July 2024
Navigating the Cyber Pandemic in Healthcare
20 June 2024
Can you see, feel, hear the difference Artificial Intelligence (AI) is making in your organisation? Does it feel real i.e. is it making a difference? There’s been much hype around AI and what it means. Every day something new seems to be happening as the pace picks up around the technology and the subtleties that go along with Artificial Intelligence. So how real is Artificial Intelligence for you and for your organisation? Firstly, to understand where you and your organisation are in the adoption of AI, let’s get to basics - what is AI?  There are many variants of definitions out there, for us at the Human Digital Collaborative, Artificial Intelligence put simply is about computers and digital devices answering an ask or solving a problem you may have by drawing their answer from a large data source. In essence, the artificial intelligence looks up menus in its library to determine quick answers for you. There are other variants of AI e.g. such as Artificial General intelligence (more human-like, sentient) which with the advent of applications Chat GPT and Claude. AI is now multi-modal i.e. text, pictures, audio, video and code can be generated from multiple inputs. This broadens the use and capabilities of AI.
20 June 2024
Welcome to the Human Digital Collaborative