What is the AI Curve and Why it's Important.
In this brave new world of rapid technological advancement, many of us find ourselves at the forefront of a monumental shift. As AI becomes increasingly integrated into the workplace and our personal lives, we must learn to navigate a landscape fraught with challenges, both technical and human.
Right now there is an amazing opportunity for those who dare, to leverage great economic wins. As demand for AI skills outweighs supply, salaries and contractor day rates will soar. Given the difficulties in our current economic times, it's wise for anyone impacted by AI to think about upskilling to ensure they become and remain relevant in the workforce.
To explain why, I will start with the patterns that occur whenever there is world-disrupting techonology.
Part One - The Curve of Adoption
Industrial Technology: What Did We Learn?
With over a century of lessons learned from technological advancements, we humans have a lot understanding and insights to draw from in terms of lessons-learnt.
Take the Industrial Revolution. While there were introductions of technology for centuries before this, it is the most prolific period in terms of large scale impact, and so seems the most logical place to start.
With it's promise of increased productivity, grand profits and improved living standards, mechanically assisted technology introduced during the industrial revolution, initially blinded people to the longer term implications. Among these includes a significant shift in employment patterns, as traditional agricultural and artisanal jobs declined while factory and industrial jobs proliferated. This shift marked the beginning of the modern workforce structure, underpinned by urbanisation and the rise of wage based labour.
But of course, with the rise in industrial machinery came unprecedented levels of air and water pollution, leading to significant and dangerous environmental degradation. From the Great Stink of 1858 and the Great Pea Soup smog in 1873, London suffered severe pollution issues as a result of poor waste management as a result of coal-fulled factory emissions and industrial pollution of the River Thames.
The longer term effects of the new technologies continued into the 20th century with the Great Fog of 1952, which saw London swamped in a thick and dangerous fog resulting in the death of thousands of residents due to exceptional levels of sulfur dioxide emitted from coal-fired power plants, factories and residential heating systems.
The introduction of technology to assist with improved productivity for textile production occurred in the mid-18th century, yet the severity of the impacts and implications of such technology on human lives wasn't addressed until the mid-20th century. These technologies, offered glittering hopes of being lifted out of poverty through solid gainful employment. But what resulted was a widening of the wealth gap with the rise of industrial capitalists owing the vast majority of assets and profits, while the lower classes in the hopes of improving their lives in the new factories, moved from rural areas to cities only to find them overrun, and with squalid living conditions.
This led to a rise in organised unions as workers began to advocate for better working conditions, along with fairer wages. Social reforms were also introduced as a means of ensuring fair treatment and equality among the classes.
The Human Cycle of Innovation.
The Industrial Revolution is an excellent example of how humans behave during the introduction of transformative technologies, and the implications they have on socio-economic factors. More or less, each technology revolution will go through these four phases with varying degrees of impact.
Innovation and Economic Boom: The introduction of new technologies often leads to economic booms, as innovative industries emerge and create wealth for those who pioneer and capitalize on these advancements. During this phase, early adopters and entrepreneurs may amass significant fortunes by leveraging the new technology to drive productivity and profitability.
Wealth Accumulation and Inequality: As the new technology matures, wealth tends to concentrate in the hands of a few individuals or corporations who control its production and distribution. This concentration of wealth can exacerbate economic inequality, as those at the top of the hierarchy accumulate vast fortunes while others struggle to compete or find meaningful employment in the changing landscape.
Social Movements and Reform: Growing disparities in wealth and living conditions often prompt social movements and calls for reform. Workers, labor unions, activists, and reformers mobilise to address the social injustices and inequities created or exacerbated by the new technology. Their efforts may lead to the implementation of labor laws, regulations, and social welfare programs aimed at mitigating the negative impacts of technological change and promoting greater economic and social justice.
Settling and Adaptation: Over time, societies tend to adapt to the new technological paradigm, integrating it into their social, economic, and political structures. Reforms and regulations may help to moderate some of the more extreme consequences of technological change, leading to a more balanced and equitable distribution of its benefits and opportunities. However, this process is often iterative and ongoing, as new technologies continue to emerge, presenting new challenges and opportunities for adaptation and reform.
Overall, this pattern underscores the dynamic and complex nature of technological innovation and its impact on society. While new technologies hold the potential to drive progress and prosperity, they also raise important questions about equity, justice, and the distribution of benefits in society, which must be addressed through collective action and social reform.
Part Two: Where Do You Fit In?
The AI Revolution
And so pausing for a second to consider it's implication and make less impulsive decision with regards to its use is, I think, the correct thing to do.
Technology leaders, are correct to harbour reservations about the implications of this particular technology, especially when there is a push for rapid AI adoption to trap early profits. The lack of education surrounding AI exacerbates this issue, as the learning curve for AI technologies is greater than perhaps a mechanical textile machine was.
For the programmers amongst you, be wary of being tasked with developing AI product feature solutions off the cuff, especially if you lack the requisite background in AI technologies. Senior management, perhaps unaware of the nuances involved, may apply undue pressure to deliver results.
It's akin to asking a PHP developer to craft a JavaScript frontend —an entirely different ballgame, despite both involving coding.
We are clearly in the Innovation and Economic boom phase, and it will not be long before massive ROI by first adopters is experienced, if not already. Like the industrial titans in their day, these new first adopters will tap into massive wealth, causing an economic boom while everyone else struggles to claw their way up the AI food chain.
If you work in a technology company or division, what sorts of things should you be thinking about just now?
For Developers
Data Processing: If you haven't seen it yet, AI can analyse and produce vast amounts of data. But of course we humans have a limit to the amount of data we can take in and process which can be overwhelming for decision makers. Training on the best methods to analysing and interpreting AI generated data in terms of accuracy and volume is a great start for developers and analysts.
Caution: Since AI comes with bias, there is no doubt going to be inaccuracies in its outputs. Understanding the nuances from the get go will put teams ahead of the game in two years time.
AI Integration: In the race to stay ahead, companies are rushing to infuse AI into their products, hoping for a competitive edge. But, as the saying goes, caveat emptor (buyer beware), as incorrect application could result in dead features and lackluster ROI.
Understanding how users interact with your product is key to unlocking its true value proposition within your product ecosystem. If you're a small team diverting resources to AI integration, it could come at the expense of other potentially lucrative endeavors.
On a positive note, I've recently encountered numerous AI-driven products that truly stand out in delivering value. Product team's who harness clear, intelligent approaches by demonstrating their commitment to serving their core clients will find their value propositions rise over those implementing AI for the sake of AI.
AI Integrations for example, can enhance efficiency in team workflows. In a recent experience, a developer attempted to use an AI tool to automate unit test writing—a task universally dreaded by developers. However, the resulting tests were deemed "total rubbish" and a waste of time. The AI output lacked the sophistication to handle edge cases, and under the pressure of looming deadlines, there was no time for reflection or analysis. Eventually, humans reverted to writing tests manually.
Whether the tool wasn't suitable for the code complexity or the developer lacked expertise in the chosen AI solution is unclear, but I'd lean towards the former, knowing the developer in question.
The key takeaway is that integrating AI into your products and workflows demands education, time and planning.
Here are some resources to start the journey.
"Hands-On Machine Learning with ML.NET" by Jarred Capellman.
"AI for Developers" by Matthew Kirk.
Pluralsight’s Machine Learning for .NET Developers
Agent Workflow AI, involves creating software agents that can autonomously execute complex workflows or tasks, often by coordinating with other agents or systems. This is an emerging disciplinein AI that has powerful benefits, so I'd strongly encourage getting in on the ground floor now.
Code reviews, CI/CD and Monitoring tasks can be enhanced with good solution setups. There is a lack of widely available training courses just now, so you will need to call in specialty services from a consultant that specialises in this area. Or you can undertake AI training for your area and try to knit it togegther.
For Leadership
Ethical Considerations: by far the biggest conversation around AI (besides how it can be monetised) is the human factor. Already in certain circles, conversations are being held to see how AI can be used to take over human tasks to increase productivity and reduce employee costs. Team leaders in particular need to adapt to the changing nature of work and find ways to leverage AI that augments human capability rather than replacing it entirely.
Also be aware that if you plan to use AI algorithms to make decisions impacting humans, such as performance reviews for examples, the technology will have inherent biases dependent on the data it was trained with. There may be inadequacies or cultural boundaries that could result in a lack of fairness, transparency and accountability in decision-making.
What is interesting with regards to the ethical considerations, is that typically these technology curves explode and disrupt markets for some time before the social reform catches up. But in this instance, we've seen the like of Elon Musk calling out the ethical issues around this technology along time before it was available to the rest of us mere mortals.
Grounded Adaptability: Grounded team leaders understand that not all changes are beneficial or aligned with the long-term goals of the company.
There must be stability to preserve morale. Even if the world is falling down around us, leaders should think carefully to formulate grounded words to communicate continuity, purpose and direction for their teams.
In the light of the ethics debate, this will help maintain momentum and anchor people in security and confidence. Decision makers must be highly empathetic in these times, understanding that the human intelligence will feel like collateral damage as the artificial intelligence rises. And rises.
If you are a team leader, then you should be seeking to skills yourself up on AI ethics, and AI business strategy to help you and your team navigate the uncertainty.
AI Policy: if your company does not have an AI policy in place with a documented strategy and approach, it would be a good idea to get one sooner than later. Careful thought and agreement should be given to how it will be used with regards to;
Sellable products: AI integrations into a product that is sold as a means of revenue.
Internal workflows: AI technology for the purpose of speeding up delivery workflows or reducing rework by decreasing defects.
Human activity: Training and education to increase skills to achieve the first two points, along with a compassionate approach to scenarios where AI takes away a task on someones job description.
Be empathetic to the fact that a programmer is probably not going to be content if all they do is review pull requests full of code generated by an AI bot. Some business models may necessitate AI to stay relevant and competitive, and that's their prerogative. And whilst fully replacing a human might not happen tomorrow, there is little doubt these decisions are around the corner so I think we need to consider how we navigate these tough new frontiers before they arrive. Regardless of the outcome, it should be done with transparency, by consultation and with compassion.
AI Going Forward
Given the cyclical nature on society transformative technology revolutions have, and the great many lessons learnt from them in the past I hope that humans have enough to stop and think ahead about the widespread implications the next wave will have.
Private enterprise will undoubtedly be at the front of the wave for obvious reasons. And I do welcome the advancements as it provides unprecedented change for the good. But like all change it will come with the bad. Changing the cycle to bring social and economic reform forward ahead of economic boom to lessen dramatic gaps in wealth would be of huge benefit for the human population.
There are two great words from antiquity, that everyone should know. One is positive and the other is negative. Behold and beware. Behold the opportunity. Behold the next person you meet might be a friend for life. Behold, look at the chances you've got. Behold the sun is shining. Behold. Here is a negative word, beware. Beware of what you become in pursuit of what you want. Don't become so obsessed in your pursuit, that it compromises your virtues, your values, the well being of those around you, and your character - Jim Rohn.
I truly believe that when it comes to AI, countries, companies and entrepreneurs, will fall into one of three categories.
There are the ones who get it right, are humane about it and flourish as a community or group. There are those who fumble with first authentic attempts and need a bit of guidance. Then, there are those who, propelled by greedy stakeholders, bulldoze their way towards AI driven dollars, with little regard for their people or indeed their own missions.
As an individual, I would encourage you to get trained in the basic principes as fast as possible to position yourself as a highly valuable employee.
You got this, go make it happen.
Authored by: Nadeen Sivic
Date: 08 May 2024
#AI #aitraining #aiagentworkflows #machinelearning #ML #MLfor.net #careers #developers #testers #CICD
Comments