Dr Frasier Crane - I'm listening
6 November 2024

Will AI ever be good enough to be a counsellor?

By Lee

Of course it will.

Look at the distance generative AI has come in the last two years. Two years ago it could barely punch its way out of a wet paper bag. Now, even professional writers and academics are admitting that generative writing AI—such as Claude or ChatGPT—have leap frogged themselves to a point where said professionals cannot write as well as the AI engines.

My mate Tim King has written an excellent piece over on Medium about how AI at the moment can’t do empathy and nuance. I totally agree with Tim, except for in the health industry (see below). But Tim is absolutely right that AI struggles with context—unless your prompt is really good, and you have plenty of support material to add to the prompt so the AI can understand at a greater level than just a prompt might achieve.

But seeing as how quickly the AI of the world has come on since two years ago, and seeing how the rush is on within the developer world to launch Agentic AI—it’s akin to a Californian or Victorian gold rush—the time is extremely near when AI will most assuredly be able to be all things to all men.

Agentic AI

Agentic AI refers to a class of artificial intelligence systems that are designed to operate autonomously, pursuing complex goals and tasks with minimal or no direct human supervision. These AI systems are capable of making decisions, planning actions, and adapting to dynamic environments based on real-time data and feedback.

Agentic AI has several key components that make it so powerful, hence why the gold rush is on to get product out:

  1. Autonomy: Agentic AI can function independently, initiating and completing tasks without constant human oversight. This allows it to handle complex workflows and decision-making processes, often in real-time environments.
  2. Goal-Oriented Behaviour: Unlike traditional AI, which typically follows predefined rules or responds to specific inputs, agentic AI is designed to pursue objectives proactively. It can set its own sub-goals and adjust strategies as needed to achieve broader targets.

  3. Reasoning and Planning: These systems use sophisticated reasoning capabilities to break down complex tasks into smaller steps (a process sometimes referred to as “chaining”). They continuously assess their progress and adapt their plans based on new information or changing circumstances.

  4. Adaptability: Agentic AI can dynamically adjust its actions in response to changes in its environment. This makes it particularly useful in scenarios where conditions are unpredictable or constantly evolving.

  5. Learning and Improvement: Through techniques like reinforcement learning, agentic AI systems can improve their performance over time by learning from their interactions with the environment.

Agentic AI is being deployed across various industries due to its ability to manage complex processes autonomously, such as:

Enterprise Automation: In business settings, agentic AI is used for workflow optimization, automating multi-step processes such as supply chain management or customer service operations.

Healthcare: Agentic AI can assist medical professionals by handling time-consuming tasks such as data analysis or treatment planning, allowing doctors to focus on patient care.

Customer Service: Intelligent personal assistants and chatbots powered by agentic AI provide personalized support by understanding user needs and responding proactively. You might remember back to the first ‘AI in Business’ podcast from AiMegos, where Gary Cooper discussed how bots could be revolutionised to actually be helpful, rather than their current frustratingly dumb state.

https://aimegos.com/ai-in-business-podcast-001

Healthcare and AI

The integration of artificial intelligence (AI) in healthcare has sparked discussions about its potential to enhance various aspects of medical practice, including the quality of patient interactions, often referred to as bedside manner. This synthesis examines whether AI has been proven to be better at bedside manner compared to human doctors.

Key Insights

AI Chatbots vs. Human Doctors in Empathy and Quality:
AI chatbots, such as ChatGPT, have been shown to provide responses that are rated significantly higher in both quality and empathy compared to those from human doctors. In a study, chatbot responses were preferred 78.6% of the time, and they were rated as empathetic or very empathetic 45.1% of the time, compared to just 4.6% for human doctors.

AI as a Complementary Tool:
AI is seen as a tool to complement, not replace, human doctors. It can handle repetitive tasks, allowing doctors to focus more on human-to-human interactions and emotional intelligence. This suggests that while AI can enhance certain aspects of patient care, the human element remains crucial.

Trust and Adoption of AI in Healthcare:
Trust in AI systems is a significant factor for their adoption by clinicians. The relationship between clinicians and AI is evolving, and trust is essential for effective collaboration. AI systems need to be designed to enhance trust and support clinicians in their decision-making processes.

AI has demonstrated a potential to provide higher quality and more empathetic responses compared to human doctors in certain settings, such as online forums.

References:

Ayers, J., Adam, P., Mark, D., Leas, E., Zechariah, Z., Jessica, B. K., Faix, D., Goodman, A., Longhurst, C., Michael, H., & David, M. S. (2023). Comparing Physician and Artificial Intelligence Chatbot Responses to Patient Questions Posted to a Public Social Media Forum. JAMA internal medicine. https://doi.org/10.1001/jamainternmed.2023.1838 

Emre, S. (2023). Artificial intelligence in healthcare: Complementing, not replacing, doctors and healthcare providers. Digital Health, 9. https://doi.org/10.1177/20552076231186520 

Fogel, A., & Kvedar, J. (2018). Artificial intelligence powers digital medicine. NPJ Digital Medicine, 1. https://doi.org/10.1038/s41746-017-0012-2 

Hamet, P., & Tremblay, J. (2017). Artificial Intelligence in Medicine. Metabolism: clinical and experimental, 69S. https://doi.org/10.1007/978-3-319-59758-4 

Jiang, F., Yong, J., Hui, Z., Yi, D., Hao, L., Sufeng, M., Yilong, W., Dong, Q., Haipeng, S., & Yongjun, W. (2017). Artificial intelligence in healthcare: past, present and future. Stroke and Vascular Neurology, 2, 230-243. https://doi.org/10.1136/svn-2017-000101 

Loh, E. (2018). Medicine and the rise of the robots: a qualitative review of recent advances of artificial intelligence in health. BMJ Leader, 2, 59-63. https://doi.org/10.1136/leader-2018-000071 

Lynn, L. A. (2019). Artificial intelligence systems for complex decision-making in acute care medicine: a review. Patient Safety in Surgery, 13. https://doi.org/10.1186/s13037-019-0188-2 

Mintz, Y., & Brodie, R. (2019). Introduction to artificial intelligence in medicine. Minimally Invasive Therapy & Allied Technologies, 28, 73-81. https://doi.org/10.1080/13645706.2019.1575882 

Onur, A., Bayrak, A. E., & Avishek, C. (2020). Artificial Intelligence and Human Trust in Healthcare: Focus on Clinicians. Journal of Medical Internet Research, 22. https://doi.org/10.2196/15154 

Silvana, S., Calandra, D., Secinaro, A., Muthurangu, V., & Biancone, P. (2021). The role of artificial intelligence in healthcare: a structured literature review. BMC Medical Informatics and Decision Making, 21. https://doi.org/10.1186/s12911-021-01488-9 

A decade hence…

In the next 5-10 years, artificial intelligence (AI) is expected to have profound and transformative effects on the Western world, reshaping industries, societies, and everyday life. Leading thinkers and technologists predict both opportunities and challenges as AI becomes more integrated into various aspects of our world.

Key Predictions from Leading Experts*

Automation and Job Transformation
One of the most immediate effects of AI will be the automation of tasks across industries. By 2030, up to 30% of hours worked in the U.S. and Europe could be automated, driven by advancements in generative AI. This will particularly affect roles in customer service, office work, and production, while demand for high-skill professions such as STEM and healthcare will rise. Millions of workers will need to transition into new roles, requiring significant reskilling efforts.

Ubiquitous AI in Daily Life
By 2030, interacting with AI will become as natural as interacting with other humans today. AI will serve as personal assistants, career counselors, therapists, accountants, and even significant others. This shift will see AI embedded in both personal and professional lives, conducting analyses, writing code, supporting customers, and making strategic decisions across industries.

Healthcare Revolution
AI is poised to revolutionise healthcare by improving diagnostics, personalising treatments based on genetic data, and enhancing accessibility through telemedicine. By 2030, AI-driven predictive analytics will allow earlier detection of diseases and more accurate predictions of patient outcomes. Wearable health monitors and virtual nursing assistants will become widespread, reducing the burden on healthcare systems and potentially leading to better patient outcomes.

Economic Inequality and Social Impact
While AI promises efficiency gains and economic growth—potentially adding $15.7 trillion to the global economy by 2030—it also risks exacerbating inequality. As AI automates tasks traditionally performed by humans, the gap between tech-savvy workers and those displaced by automation may widen. Some experts suggest that without careful management of these transitions (e.g., through Universal Basic Income or new economic models), social unrest could increase due to rising inequality.

Accelerated Pace of Life
AI will make decision-making faster for businesses, governments, and individuals alike. This will lead to a sense that life is speeding up as organisations use AI to handle complex decisions more quickly. In education, AI will tailor learning experiences to individual students’ needs, while in finance and law, it will streamline processes such as fraud detection or legal contract drafting.

Projections That May Surprise You

Emotional AI and Human-AI Relationships
By 2030, it is predicted that humans may form deep emotional bonds with AI systems. These relationships could go beyond professional interactions; some experts believe it will be commonplace for humans to have AIs as significant others. Additionally, advancements in emotion-reading technologies could enable AIs to personalise customer experiences based on real-time emotional feedback.

AI-Driven Environmental Conservation
AI is expected to play a pivotal role in environmental conservation efforts by 2030. Smart drones equipped with AI will monitor endangered species and track poachers, while AI algorithms optimise resource usage in cities to reduce waste. These systems could help address climate change by improving sustainability practices across industries.

Quantum Computing’s Role in AI Advancements
Some experts believe that quantum computing could be critical for achieving human-level or superhuman intelligence in AI systems within the next decade. Quantum computers’ ability to process vast amounts of data at unprecedented speeds may unlock new capabilities for AI that are currently unimaginable.

Autonomous Vehicles Becoming the Norm
Self-driving cars are expected to dominate transportation by 2030. Autonomous vehicles (AVs) will communicate with each other to optimise traffic flow and reduce accidents. This shift could render human drivers obsolete in many contexts—transforming not only personal transportation but also logistics and delivery services.

New Industries Focused on Human Interaction
As automation takes over many tasks traditionally performed by humans, there may be a growing demand for industries centred around human interaction. Some experts predict a backlash against technology-driven systems like chatbots or automated customer service lines, leading to a resurgence of businesses offering genuine human contact.

Conclusion

The next 5-10 years will see rapid advancements in artificial intelligence that transform many aspects of life in Western countries—from how we work and receive healthcare to how we interact with machines daily. While these changes offer significant opportunities for efficiency gains and innovation, they also present challenges around job displacement, inequality, privacy concerns, and ethical dilemmas.
Leading thinkers urge caution as society navigates this new landscape: ensuring that AI development is guided by ethical considerations and that its benefits are shared broadly across society will be crucial for a positive future shaped by artificial intelligence.

Already, generative wordsmiths like Claude and ChatGPT can produce better content than the vast majority of writers and copywriters out there. What used to take me 3-4 days to produce—such as a communication piece for a client, complete with their ‘voice’, now takes Claude or ChatGPT less than a minute, and if the prompt is good enough, in the client’s ‘voice’ and ‘tone’.

I’ve already written several books that have gathered comments such as:

Except I didn’t write all of them, Claude did, from a ChatGPT outline. I just used ProWritingAid to tidy them up. They were already in my ‘voice’ and style. The writer-purists may scoff and tut-tut, but my Amazon author page is awash with books, and that does my profile no harm at all. My first social media books took me 2-3 months to write; my first novel took three years (and it was so atrociously bad not even my closest friends would finish it). Now I can publish a book every month, sometimes every two weeks. I loves AI.

*Experts:

Daron Acemoglu
• Affiliation: MIT Professor
• Insights: Acemoglu has discussed the potential risks of AI deployment, particularly regarding worker protections and prosperity. He argues that while AI can bring innovation and prosperity, its rapid deployment by large tech corporations may underestimate the value of human workers.
• Source: Inequality in the Digital Age conference

Jean-Marc Laouchez
• Affiliation: President, The Korn Ferry Institute
• Insights: Laouchez emphasizes that while AI will automate many tasks, humans will remain central to decision-making processes. He highlights the importance of integrating AI to boost productivity and operational efficiency while maintaining a focus on human capital.
• Source: Korn Ferry

Cynthia Breazeal
• Affiliation: Professor at MIT, Director of the Personal Robots Group
• Insights: Breazeal focuses on social robotics and the long-term impact of AI on everyday life. She advocates for AI systems that enhance human flourishing by supporting social-emotional intelligence and collaboration.
• Source: Success Magazine

Demis Hassabis
• Affiliation: CEO of Google DeepMind
• Insights: Hassabis discusses the future breakthroughs in AI, emphasizing that advancements will require more than just scaling current technologies. His work at DeepMind involves developing innovative AI models like Gemini.
• Source: WIRED

Jaroslaw Rzepecki
• Affiliation: CTO at Monumo
• Insights: Rzepecki notes that AI will lead to shifts in job roles, with some being replaced and new ones created. He believes that jobs requiring manual dexterity or high creativity will prevail in the short to medium term.
• Source: Tech Funding News

These experts provide a range of perspectives on how AI is likely to influence work and employment, highlighting both opportunities for innovation and challenges related to job displacement and economic inequality.