This Week in Venture Capital and Artificial Intelligence: Knock Knock, Who’s There? AI

Every week I will share and summarize ideas and insights with the LinkedIn community. Some of the summarizations I may have covered during the week, and other things may have just been passing thoughts.

You can find out more about me by tuning into my LinkedIn profile daily and peaking my CV that is listed below from time to time. Also if you would like to be a guest on my 2 podcasts, Capital Unscripted, or Explainable AI, or if you would like to be a contributor to any of the medias I write for or am partnered with, -- includes AI Accelerator Institute , Idea Scale, AI news or other well-known media outlets such as Yahoo Finance -- then please reach me by inboxing me. CV: www.paulclaxton.io

Read to the end for this terrible, terrible joke (LoL)…

It is not necessarily the existence of AI we need or necessarily a needed understanding of what AI is, but it is an understanding of the other things we need that AI makes better.

We can’t efficiently leverage AI to improve our lives, work and our world until we have an omnipresent human understanding to know exactly what AI does. In other words, we cannot totally understand what AI is, until we know what it does. 

I am going to prove to you that we have no synonymous and consistent understanding as humans as to what AI actually is because we don’t quite know exactly what it does.

Case In Point: We know what air does, but many people could not tell you exactly the chemicals and composition that make up air. And truthfully, we do not need to know what makes up air, even though we all mostly have some general idea, and we should have a general idea of what makes up air, however, what is most important is that we know what air does.

Most people will tell you that air is full of oxygen. Most people will tell you water is comprised of H₂O" which is "two hydrogen atoms and one oxygen atom. While this is true, many people may forget the other components that make each possible as well as how they are formed.

We know what air does, even if many people cannot tell you its exact chemical compositions. While most people understand that air is imperative for breathing and that water is imperative for hydration, they may not know that air is primarily made up of nitrogen, not oxygen as many would associate with air, or that water is formed when hydrogen and oxygen come together in a chemical reaction, long after you have turned on your kitchen faucet, or when it rains!! The water we see today has already been in the Earth’s water cycle for many years, which is very similar to AI in that the data that powers AI has been in human generations for years, getting recycled and reused over and over again.

Here is my ask to random people whom I have asked to answer objectively, regardless of how much they know about me or how well they know me. I did not ask them to tell me their own description of AI regardless of how they feel about it, because each answer should come from the heart. So if they choose to answer objectively, or subjectively based on their own predisposition to AI, then that is their choice, and we know we are getting the real answers.

“At the moment, I am trying to gain different insights from individuals to be used in confidentiality for an article I am writing. I am asking a handful of people to describe what they think AI is their own words. Would you be open to giving me your own description of what you think AI is? A paragraph description, more or less, should suffice.”

Chat - GPT Responder #1:

Artificial Intelligence (AI) is a field of technology that enables machines to perform cognitive functions similar to human capabilities, such as learning, problem-solving, and decision-making. AI leverages algorithms and vast data to identify patterns, make predictions, and automate tasks that typically require human intelligence. As a powerful tool, AI offers numerous benefits, including enhanced productivity, personalized experiences, and advancements in areas like healthcare and education. However, its effectiveness and output are heavily influenced by human input, which can introduce biases and shape its applications.

Despite its impressive capabilities, current AI systems have limitations. They often excel in narrow, task-specific applications but lack the generalized intelligence to adapt across different domains. The potential for AI’s future, however, is vast, as ongoing research aims to develop more adaptable and context-aware systems. While AI holds promise for solving complex problems and driving innovation, its drawbacks—such as ethical concerns, job displacement, and privacy issues—underscore the need for responsible development and oversight. 

Human Responder #2:

AI is a tool I use to work smarter. It functions as both a sense maker and aggregator of data. As a writer, I use AI for grammatical corrections, meeting note transcription, thematic analysis. I use ChatGPT to aggregate general information about case studies and then I do further research to filter case examples through my visionary leadership model. AI has also been helpful in paring down my final drafts as I tend to overwrite (lol). I also prompt ChatGPT to revise cold emails to bypass the Google/GMass spam filter, while preserving my original tone and voice. AI is a boon to business for those willing to invest the time to use it. Like in any space that functions as the “Wild West,” the user has to harness its power to reap the greatest benefits.

Human Responder #3:

It has the same critical failing of social media - creating an echo chamber of dopamine by reinforcing my every comment or prompt with how smart, insightful and dynamic I am. (I am not as good as AI chat claims I am). This echo chamber is the death of discourse in our world. The challenge is no one is looking at the implications for our society, our economy, our politics or relationships. They are distracted by a false narrative of world domination and sentience.  The real terror is when we all stop thinking critically about ourselves, our ideas and opinions and our relationships and responsibilities to our society and our freedoms. We have a precursor in social media echo chambers.  What happens when it's reinforced by AI?  This is the true issue with bias... the bias we bring to the conversation.

Human Responder #4:

My reactive response for AI. The movie ‘terminator’

My proactive response. AI is a tool, nothing more, can be used for good, like helping find a specific location or specific file that sometimes an individual can’t quite remember it exactly, or used to harm, like creating a voice to mimic an individual to a certain reaction or response.

Human Responder #5:

AI often gets cast as a "villain" in mainstream media, and society as a whole, instilling fears about job loss or control. Kinda like every movie must have a bad guy to blame for all of the mistakes. A robot (the ones we see for example in movies causing destruction) is just the vessel, it is not the AI. AI output has to have an input. That input comes from ML (machine learning). That ML learns from the HUMAN putting the information INTO the ML, which then becomes the AI output. In other words, Bob may have a biased opinion on XYZ, so he INPUTS his biased info, which OUTPUTS as biased, and poor wee Mr./Mrs. AI robot gets the blame.

We need to get more stringent on who we allow to put the information in (the humans)!

People are always naturally scared of what they don’t really understand. Even though we have a lot to learn about AI, including how to effectively regulate it in a way that doesn’t suffocate innovation, when we take a closer look objectively, AI's real-world applications have already made a massive, positive impact.

In education, AI personalizes learning, offering students tailored resources and support to suit their learning styles (yes that is actually a thing), especially for people like me that are neurodiverse. This was not available in my school days in the early 90’s and would have been very beneficial as a neurodiverse student. AI also supports my verbal thought process and helps me construct my own information/thoughts. In healthcare, it speeds up diagnosis with much more accuracy, and treatment planning, saving lives. The first AI assisted brain surgery has also just been successfully completed with more precision that ever before!

Automation might take over some repetitive tasks (which nobody likes doing anyway), but it also frees up people to focus on more creative, strategic, and impactful work. Rather than just replacing jobs, AI transforms them, bringing new opportunities for skill-building and efficiency, adding volume to peoples’ skillsets that they might not otherwise have even known they had.

The key is to be open-minded, have a responsive (not reactive) non-biased attitude to make informed decisions, harness AI as a tool that elevates human potential rather than replacing it, and guiding it in ways that benefit society.

Human Responder #6:

The question you ask is not easy or trivial as the definition of AI could, and has, really filled books. It’s everything from simple decision-making based on analytics programmed by humans to what is called generalized artificial intelligence. Such generalized AI world mimic the human brain in that it can apply intelligence in any field of knowledge. I tend to think of the things that are being done today as being the relatively simple kind. That is, the intelligence which seems really smart, is really based on algorithms that are known and input to the AI engine by people.  Being able to ferret out useful information from mass amounts of data is not intelligence... it's just better data processing that's been going on since the creation of computers (and I include analog computers). Today AI is only being applied in fairly narrow domains. Despite the hype, it is not revolutionary... it's just incremental extension of what's been going on for decades.

 When I think of the more interesting forms of AI, generalized artificial intelligence, I tend to think about artificial intelligence that is not pre-programmed decision making defined by humans, but an intelligence that can look at data, look at the world and create unexpected information and unanticipated insights that has not been pre-programmed by the "founding fathers."

AI today is also limited in that much of the days you'd want to build intelligence upon is proprietary. To create an extraordinary generalized, your world need access to vast amounts of diverse data and algorithms that are able to learn and create new algorithms of their own.  I think that form of generalized artificial intelligence is still a few years away, if not longer, but it is coming.

There is much more to say and one risks trivializing the subject by addressing it in sound bites and opinions (including my own).

It is clear we have no consistent understanding of what AI is, because everyone’s answer travels from the heavens to hell on AI, with each answer having many biases, inconsistencies and contradictions to the previous responder’s answer.

Let’s Explore The Inconsistencies and Contradictions:

Understanding of AI:

• Responder #2 and #5 view AI as a practical tool with clear benefits for productivity, business, education, and healthcare. They highlight how AI can empower users to work smarter and more efficiently.
• Responder #3 paints AI as a potentially dangerous echo chamber, similar to social media, reinforcing biases and eroding critical thinking.
• Responder #6 argues that current AI is not revolutionary but simply an extension of traditional data processing, emphasizing the gap between narrow AI and the promise of generalized AI.
• Responder #4 holds a simplistic and fear-based view rooted in popular culture, seeing AI as both a useful tool and a potential threat reminiscent of movies like The Terminator.

Portrayal of AI's Impact:

• Responders #2 and #5 focus on AI's positive, transformative potential, especially in facilitating creativity and supporting various industries. Responder #3 underscores the negative implications, likening AI to a societal disruptor that hinders discourse and critical reflection. Responder #6 argues that current AI, despite its advancements, is not revolutionary and emphasizes that real AI intelligence—generalized AI capable of independent and innovative thinking—is still on the horizon.

Human Agency and Control:

• Responder #5 emphasizes that human input drives AI's output, suggesting that AI's impact depends heavily on the users who shape its behavior. This implies a level of human control that Responder #3 overlooks when describing AI as an amplifier of human bias. Responder #6 suggests that current AI lacks true autonomous intelligence and is limited to what humans program into it, which contrasts with Responder #3's fears of AI as a societal threat.

Role of Bias and Regulation:

• Responder #3 is concerned with AI's reinforcement of human bias and its potential to create echo chambers. Responder #5 acknowledges human bias as part of AI's output but focuses on how society can regulate and harness AI for good. Responder #2 does not mention bias directly, focusing instead on practical uses, whereas Responder #4 hints at potential misuse but without deep analysis.

How These Contradictions Could Create Indifference:

• Diverse and Conflicting Narratives. The varied descriptions contribute to an inconsistent narrative about what AI truly is and what its capabilities and limitations are. For some, AI is a useful tool that needs careful handling and regulation (Responder #5). For others, it is a source of existential concern (Responder #3), a mundane extension of computing (Responder #6), or a misunderstood movie villain (Responder #4). These conflicting views can confuse individuals and lead to cognitive dissonance.
• Fragmented Understanding. The lack of consensus means people may cherry-pick the interpretation that aligns with their existing beliefs or ignore the subject entirely due to its perceived complexity or overhyped nature. If AI is seen simultaneously as an indispensable tool, a harbinger of social discord, an overblown continuation of existing technologies, and a misunderstood antagonist, the true, nuanced understanding of its potential and limitations gets lost.
• Public Apathy and Misinformation. When faced with such a wide range of views, the public may become indifferent, dismissing the topic as too complicated or contradictory to form a clear stance. This indifference can prevent individuals from fully engaging with discussions about AI's ethical, social, and economic impacts or from understanding its role in shaping the future.
• Impacts on Adoption and Policy. Conflicting opinions can stall both AI adoption and the development of coherent policies. Some may embrace AI for its benefits, while others, influenced by fears of bias or societal impacts, may resist or push for restrictive measures that hamper innovation. Without a unified perspective, decision-making at the societal level becomes fragmented, and responses to AI's challenges may be piecemeal or ineffective.

The very real and present contradictions we have explored here from real people of various astute backgrounds in how AI is described—ranging from a helpful tool to a societal disruptor or an overhyped extension of computing—can lead to public confusion, indifference, and polarized opinions that shape perceptions and engagement with AI unevenly.

Indifference. And this is the problem we have.

Truths and Extremities In Bias:

Each view on AI captures elements of truth but is limited by biases and extremities. Responder #2 accurately presents AI as a productivity tool but does not focus much on the complexities such as biases and ethical concerns. Responder #3 correctly highlights how AI can reinforce biases and create echo chambers but leans pessimistic, overlooking positive uses and bias-mitigation efforts. Responder #4’s pop culture-based view acknowledges potential risks but dramatizes AI, distracting from its real capabilities. Responder #5 rightly discusses human input shaping AI outcomes but downplays challenges like job displacement and privacy. Responder #6 is realistic in describing current AI as incremental but may under-appreciate its transformative impact.

These perspectives stem from individuals of a wide pool from ages 22 years old to 70 years old, including business owners, corporate decision-makers, and those with political influence—amplifying the significance of differing opinions. This variance makes AI a hard to manage or regulate volatility and a very real threat based on our misinterpretations and differing views, sparking both hope and concern in many, as it embodies potential and pitfalls that resonate differently across various ages, sectors and levels of power.

These views collectively paint a kaleidoscopic picture of AI, each with its own positives and drawbacks. The most balanced understanding of AI’s role was Chat-GPT’s which incorporated it as a powerful tool with benefits and drawbacks, and acknowledged the influence of human input, and recognized both current limitations and future potential.

I want to highlight Responder #5’s statement: “A robot (the ones we see for example in movies causing destruction) is just the vessel, it is not the AI. AI output has to have an input.” And that really is the key is the input that we collect from some of the conflicting views from billions of people across the world using AI.

This makes clear the critical point that the input we gather, shaped by the varied and conflicting perspectives of billions of people, is what drives AI’s behavior. This is why I believe that if we use AI to shape our understanding of what AI is, and identify and reduce bias in the inputs, then AI itself would reflect less bias.

Efforts are already underway with bias-detection algorithms and fairness frameworks, but eliminating bias entirely is unrealistic due to the law of nature that there has to be a winner and a loser, a predator and prey, a fight or flight, etc.

Confirmation Bias and the Tomato:

Tomato, tomayto. Whether a tomato is a fruit or a vegetable relies on the regulated definition of a tomato. But no one goes by that. They go by their own bias. I just asked my girlfriend and she thinks it is a vegetable. I would agree. But the botanic definition is that it is a fruit. Her answer and mine too are simply biased, “I simply just view it as a vegetable,” she says…

What makes either person right or wrong is not the botanic definition, it is our own view, even if it is technically incorrect. So we cannot agree, despite the obvious. This is the age old debate, and every one seems to have a different opinion about what a tomato is and no one agrees universally. The result means that a tomato get used in ways that may be outside the traditional fruit usage category.

For example, in culinary practices, people who view tomatoes as vegetables many times use them in savory dishes such as salads, soups, and sauces, where they complement other vegetables and create flavor profiles associated with main courses or side dishes. This perception aligns with their categorization in grocery stores and recipe books, where they are commonly found alongside vegetables.

Conversely, understanding tomatoes as fruits—based on their botanical definition—can lead to their inclusion in contexts typically reserved for fruits, such as fruit salads or even desserts, albeit less commonly. This broader perspective can inspire creative uses that cross traditional culinary boundary, showcasing the tomato’s versatility.

In broader terms, different perspectives can influence commercial practices, agricultural classifications, and dietary guidelines. For example, tariffs and trade regulations have historically classified tomatoes based on their common use as vegetables, affecting how they are imported and taxed. Thus, the understanding of whether a tomato is a fruit or a vegetable shapes both its cultural identity and practical applications.

We are all sitting here debating about what to do with AI, but none of us have a consistent understanding of what AI is and you can’t debate how to manage something when the other person has a completely different idea of what that something is in existence and functionality; it’s like an English speaking person and a Japanese person trying to debate but they’re speaking totally different languages and cultures.

For example, consider a panel with an AI researcher, tech executive, policymaker, and journalist. The researcher views AI as task-focused algorithms, the executive sees it as a tool for innovation, the policymaker focuses on societal impacts, and the journalist frames it with public fears and risks. Without a shared understanding, their debate on AI governance becomes fragmented, hindering cohesive decision-making. This illustrates the need for a consistent baseline to have meaningful discussions about managing AI.

Or consider how people view airplanes. One person might see an airplane as a people mover, another as a cargo carrier, and yet another as a military tool. Each perspective leads to different regulations and priorities—passenger safety standards, cargo handling protocols, or defense strategies. Similarly, in AI, if people hold varied perspectives—one viewing AI as simple automation, another as a business tool, and another as a societal risk—debates on governance will be fragmented as they are now.

If we operated the airplane industry the way we currently approach the AI industry, we would face many more challenges and inefficiencies. The airplane industry would lack consistent global standards, resulting in fragmented regulations shaped by each stakeholder’s differing views.

For instance, passenger airlines might focus solely on safety and comfort without addressing cargo handling or military security needs. Cargo operators would prioritize logistics, potentially overlooking passenger safety or defense strategies, while military stakeholders might set their own stringent protocols without considering commercial implications.

This fragmented approach would lead to confusion, safety risks, and inefficiencies, as regulations and practices would vary widely based on subjective perspectives. Coordination between nations and industries would be difficult, slowing down progress and causing conflicts over priorities. Without a unifying body or consistent framework to align all stakeholders, the industry would struggle to maintain safety, innovation, and efficiency. In the context of AI, similar fragmented approaches result in unclear standards, varying levels of risk management, and challenges in ethical governance, making important the need for a comprehensive, international unified understanding, approach and strategy.

We should be using AI majorly as an objectivity tool. We are using AI in the wrong way in that we are still relying largely on human bias and to proliferate human weaknesses such as laziness, and we are also using AI to perpetuate this bias, instead of looking for the answers we do not want to hear, the truth. Similar to Responder #3’s statement, “It has the same critical failing of social media - creating an echo chamber of dopamine by reinforcing my every comment or prompt with how smart, insightful and dynamic I am.”

 I propose that every important decision affecting livelihoods, privacy, and human welfare should be run through AI as a means to minimize bias and promote balanced, informed outcomes.

 Rely On AI Not The Other Guy:

Relying on AI to help us understand AI instead of relying on ourselves to understand it is inherently ironic, as it creates a loop where we depend on a tool to interpret itself. This feels illogical because AI, lacking human emotion or ethical judgment, is asked to explain concepts embedded deeply in human society and identity yet also take many differing views and provide balanced insights. Unlike human interpretations, which are many times biased and fragmented, AI can help us reduce these biases, adapt in real time to new developments, and provide an evolving overview that no single human or group could achieve alone. This self-referential relationship makes it clear both the limitations of human perspectives and the important role AI plays as both subject and teacher, facilitating a more complete understanding of its own nature and impact.

The only way to successfully regulate AI I believe will be to find a general consensus and agreement to what AI is, not what it can do, but what it is and regulate, reward and penalize its use. Because anyone across the world in their basement can build an algorithm, and anyone can get data, anyone can build AI, so how do you regulate who builds what across the world and how it is built. You can’t. But what you can do is regulate rewards and penalties for its use. Similar to how nuclear technology is regulated: While a nuclear weapon may be a much tougher feat, many countries and vigilante groups can build a bomb of massive scale. So it’s difficult to control who learns about nuclear science or conducts theoretical research, the use of nuclear technology is tightly regulated, with clear rewards for peaceful applications (e.g., energy) and severe penalties for weaponization or unsafe use. A similar approach could work for AI, focusing on how it’s applied and holding users accountable to established ethical and legal standards.

Using AI to find a general consensus among differing viewpoints on debatable topics concerning the use, production, and deployment of AI is something that should be exercised much more than it has been…

This approach can help clarify complex debates by organizing viewpoints and facilitating more objective discussions. It should also serve as a tool for policymakers, researchers, and stakeholders to make informed decisions based on a comprehensive understanding of the collective sentiment and knowledge landscape.

Conclusion:

The varied perspectives shared in this article underline a critical truth: without a unified understanding of what AI truly is, discussions about its regulation, impact, and ethical use will remain fragmented and unproductive. Just as debates about whether a tomato is a fruit or a vegetable influence its cultural and practical use, differing views on AI shape our engagement and strategies. The challenge is not just defining what AI can do but building consensus on what AI is at its core—a tool shaped by human input, carrying both promise and peril. To navigate this complexity, leveraging AI itself as a facilitator of balanced, comprehensive analysis could bridge the gaps in human bias and fragmented opinion. Ultimately, generating a shared, objective framework for understanding AI is essential for effective regulation, responsible development, and maximizing its potential to enhance human welfare while minimizing risks. This collective clarity is the first step toward scaling AI not just as a tool but as a partner in shaping a better future.

Because AI is such a sensitive and controversial subject, I have not redacted any responder’s viewpoints or altered their words and so I have kept names silent just to keep a bit of indifference to the views supported here, I want to thank each individual for their contributions to this article, without calling out which responder they are directly. I have shared this article with each responder, and for those responders that so generously contributed to this article, if you would like to claim your response, please reshare this article claiming your particular responder number. 

Knock Knock, Who’s There? AI!

Ehhhh, I make your decisions from now on!

You can find out more about me by tuning into my LinkedIn profile daily and peaking my CV that is listed below from time to time. Also if you would like to be a guest on my 2 podcasts, Capital Unscripted, or Explainable AI, or if you would like to be a contributor to any of the medias I write for or am partnered with, -- includes AI Accelerator Institute , Idea Scale, AI news or other well-known media outlets such as Yahoo Finance -- then please reach me by inboxing me.

I hope you enjoyed this week's newsletter stay tuned for next Saturday's edition.

How to contact me:

-- Other than LinkedIn, if you want to know more about me or hear more from me you can view my CV here: www.paulclaxton.io

-- You can also schedule a meeting with me here by going to the bottom of my business card and following the instructions. Business card

#ai #artificialintelligence #vc #venturecapital #entrepreneurship #leadership #leaders #aiblogs #vcblogs #venturecapitalist #founders #aipolicy #paulclaxton   #innovation #aibias #sustainableai #aidecisions #aidecisionmaking #futuretech #airegulation #aigovernance