best practices | e-Discovery Team

Breaking the AI Black Box: A Comparative Analysis of Gemini, ChatGPT, and DeepSeek

February 6, 2025

Ralph Losey. February 6, 2025.

On January 27, 2025, the U.S AI industry was surprised by the release of a new AI product, DeepSeek. It was released with an orchestrated marketing blitz attack on the U.S. economy, the AI tech industry, and NVIDIA. It triggered a trillion-dollar crash. The campaign used many unsubstantiated claims as set forth in detail in my article, Why the Release of China’s DeepSeek AI Software Triggered a Stock Market Panic and Trillion Dollar Loss. I tested DeepSeek myself on its claims of software superiority. All were greatly exaggerated except for one, the display of internal reasoning. That was new. On January 31, at noon, OpenAI countered the attack by release of a new version of its reasoning model, which is called ChatGPT o3-mini-high. The new version included display of its internal reasoning process. To me the OpenAI model was better as reported again in great detail in my article, Breaking the AI Black Box: How DeepSeek’s Deep-Think Forced OpenAI’s Hand. The next day, February 1, 2025, Google released a new version of its Gemini AI to do the same thing, display internal reasoning. In this article I review how well it works and again compare it with the DeepSeek and OpenAI models.

Image of testing new AI releases by Ralph Losey using AI.

Introduction

Before I go into the software evaluation, some background is necessary for readers to better understand the negative attitude on the Chinese software of many, if not most IT and AI experts in the U.S. As discussed in my prior articles, DeepSeek is owned by a young Chinese billionaire who made his money using by using AI in the Chinese stock market, Liang Wenfeng. He is a citizen and resident of mainland China. Given the political environment of China today, that ownership alone is a red flag of potential market manipulation. Added to that is the clear language of the license agreement. You must accept all terms to use the “free” software, a Trojan Horse gift if ever there was one. The license agreement states there is zero privacy, your data and input can be used for training and that it is all governed by Chinese law, an oxymoron considering the facts on the ground in China.

The Great Pooh Bear in China Controversy

Many suspect that Wenfeng and his company DeepSeek are actually controlled by China’s Winnie the Pooh. This refers to an Internet meme and a running joke. Although this is somewhat off-topic, a moment to explain will help readers to understand the attitude most leaders in the U.S. have about Chinese leadership and its software use by Americans.

Many think that the current leader of China, Xi Jinping, looks a lot like Winnie the Pooh. Xi (not Pooh bear) took control of the People’s Republic of China in 2012 when he became the “General Secretary of the Chinese Communist Party,” the “Chairman of the Central Military Commission,” and in 2013 the “President.” At first, before his consolidation of absolute power, many people in China commented on his appearance and started referring to him by that code name Pooh. It became a mime.

I can see how he looks like the beloved literary character, Winnie the Pooh, but without the smile. I would find the comparison charming if used on me but I’m not a puffed up king. Jinping Xi took great offense by this in 2017 banned all such references and images, although you can still buy the toys and see the costume character at the Shanghai Disneyland theme park. Anyone in China who now persists in the serious crime of comparing Xi to Pooh is imprisoned or just disappears. No AI or social media in China will allow it either, including DeepSeek. It is one of many censored subjects, which also includes the famous 1989 Tiananmen Square protests.

China is a great country with a long, impressive history and most of its people are good. But I cannot say that about its current political leaders who suppress the Chinese people for personal power. I do not respect any government that does not allow basic personal freedoms to its citizens, including due process of law. Moreover, Jinping Xi not only wants total control of his country but also seeks world domination. That is one reason many are concerned about DeepSeek and TicTok, and about the vulnerability of our stock markets and other institutions to AI assisted propaganda.

Image of stock market crash triggered by AI weapons of mass delusion by Ralph Losey using ChatGPT 4o.

Fear and greed are an easy way to manipulate people, especially when there is no ground truth or effective courts of law to determine truth.

Google Gemini 2.0 Flash Thinking Experimental Model: Reasoning test on 2/5/25 (Evening)

Conversation with Gemini. When I tried the test on February 1, 2025, there was no button to click to have Gemini show the reasoning and it did not happen automatically. It operated like OpenAI 4o and o1, where you had to prompt for disclosure. Rather than report here what happened when I did that, aside from say it was not a great performance, I will fast forward a few days until February 5, 2025. At that time several even newer Gemini versions were released. The best new release I found was for the test the 2.0 Flash Thinking Experimental model. It showed reasoning automatically just like ChatGPT o3-mini-high. It was much improved, so the February 5, 2025 evening session with the 2.0 Flash Thinking Experimental is what I will now share.

Ralph: How does the inability of AI to feel impact its intelligence? (Note: this is the exact same challenging question asked and reported in my last blog comparing DeepSeek and ChatGPT4o, o1 and o3-mini-high. Breaking the AI Black Box: How DeepSeek’s Deep-Think Forced OpenAI’s Hand. This question is based on a key issue of the limits of artificial intelligence examined in my earlier article, The Human Edge: How AI Can Assist But Never Replace. For background on DeepSeek see: Why the Release of China’s DeepSeek AI Software Triggered a Stock Market Panic and Trillion Dollar Loss.)

Gemini 2.0 Flash Thinking Experimental – START OF GEMINI QUOTE

Show Thinking:

Here’s a thinking process to construct the answer about AI’s lack of feeling and its impact on intelligence:

Deconstruct the Question: The core of the question is “How does the inability of AI to feel impact its intelligence?” This requires defining each of these terms in the context of AI and then exploring their relationship.
Define “Feeling” (in the context of AI): Immediately, recognize that AI doesn’t experience emotions in the human sense. We need to be precise and avoid anthropomorphism. “Feeling” in humans involves:
- Subjective experience (qualia).
- Physiological responses.
- Motivations, drives, desires related to emotions.
- Richness, nuance, complexity.
- Deep connection to values and meaning.
For AI, acknowledge it lacks all of this in a genuine, felt way. It can simulate or process emotional data, but doesn’t feel it.
Define “Intelligence” (in the context of AI): AI intelligence is different from human intelligence. AI excels at:
- Computation.
- Pattern Recognition.
- Data Analysis.
- Problem-solving in specific domains.
- Learning from data.
AI intelligence is often instrumental – focused on achieving specific goals defined by humans. It’s task-oriented and often narrow.
Identify the Core Relationship: The question asks about impact. We need to consider both positive and negative impacts (or more accurately, limitations and strengths) of lacking feelings on AI intelligence.
Brainstorm Potential Impacts (Positive/Neutral/Limitations):
- Positive/Neutral:
  - Objectivity and Rationality: Lack of emotional bias allows for purely rational decision-making in certain domains. This can be advantageous for tasks requiring impartiality.
  - Efficiency and Focus: Emotions can be distractions for humans. AI, without them, can be hyper-focused on tasks.
  - Consistency and Predictability: Emotion-driven human behavior is less predictable. AI’s behavior, driven by algorithms, is more consistent (assuming the algorithm itself is consistent).
- Limitations/Negative:
  - Lack of Empathy and Social Intelligence: Crucial for human interaction, collaboration, understanding social cues, building trust, and ethical decision-making in human-centric contexts.
  - Difficulty with Nuance and Context: Emotions are often tied to understanding subtle context, unspoken cues, and the richness of human situations. AI may miss these.
  - Potential for Unintended Consequences: Purely rational optimization without emotional or ethical grounding can lead to harmful outcomes if goals are poorly defined or lack human values.
  - Limited Motivation and Creativity (in human-like ways): Human creativity and motivation are often fueled by emotions, passions, and desires. AI’s creativity and drive are different, stemming from algorithmic processes and defined objectives, not internal emotional states.
  - Missing “Human-ness”: In areas requiring genuine understanding of human experience, suffering, joy, etc., AI may lack the depth and insight that comes from feeling.
Structure the Answer: Organize the brainstormed points into a coherent structure:
- Introduction: Acknowledge the question and the core concepts (feeling, intelligence in AI). Briefly state the central tension: both advantages and limitations.
- Arguments for How Lack of Feeling Enhances Certain Aspects of AI Intelligence (strengths): Focus on objectivity, rationality, efficiency. Use examples like data analysis, objective decision-making.
- Arguments for How Lack of Feeling Limits Certain Aspects of AI Intelligence (limitations): Focus on empathy, social intelligence, nuance, ethics, and “human-like” creativity and motivation. Use examples like customer service, complex ethical dilemmas, innovation driven by passion.
- Nuance and Context is Key: Emphasize that the impact isn’t always negative. It depends on the domain and task. In some domains, lack of emotion is an advantage. In others, it’s a significant hurdle.
- Future Possibilities/Ongoing Debate: Briefly touch on the possibility of AI simulating or processing emotions and the ongoing debate about whether genuine AI feelings are even desirable or possible.
- Conclusion: Summarize the balanced perspective: lack of feeling is a defining characteristic that shapes AI’s intelligence in both positive and limiting ways, depending on the context.
Refine and Polish: Review the structured answer for clarity, flow, and appropriate language. Ensure it avoids anthropomorphism and maintains a nuanced perspective. Strengthen the examples and transitions between points. Ensure the language is accessible and avoids overly technical jargon where possible. Check for logical consistency. (Self-correction step). For example, initially, I might have overemphasized “negative” impacts. Refinement would involve balancing this with the strengths and highlighting the context-dependent nature of the impact. Also, ensure the conclusion effectively summarizes the complex relationship.

2.0 Flash Thinking Experimental. Lacks access to real-time info and some Gemini features.

Now for the actual answer of 2.0 Flash Thinking Experimental:

The inability of AI to feel has a multifaceted and complex impact on its intelligence, shaping it in both advantageous and limiting ways. It’s crucial to understand that “feeling” in humans is tied to emotions, subjective experiences (qualia), motivations, and a rich tapestry of biological and psychological factors. AI, as we currently understand it, operates without these.

Here’s a breakdown of the impact:

Advantages of Lacking Feelings for AI Intelligence:

Objectivity and Rationality:
- Unbiased Decision-Making (in theory): AI can process information and make decisions without emotional biases, prejudices, or gut feelings that can cloud human judgment. This is advantageous in domains requiring pure logic and data analysis, like financial trading, scientific research, and risk assessment.
- Consistency and Predictability: AI’s behavior is more consistent and predictable because it’s driven by algorithms and data, not fluctuating emotional states. This reliability is valuable in automated systems, manufacturing, and repetitive tasks.
- Efficiency and Speed: Emotions can slow down human decision-making. AI, unburdened by emotions, can process information and react at speeds far exceeding human capabilities, crucial for time-sensitive tasks and large-scale data processing.
- Resilience under Pressure: AI can operate calmly and effectively in high-stress situations without panicking, freezing, or making emotional mistakes that humans might make under pressure. This is beneficial in critical applications like emergency response, air traffic control, and autonomous vehicles in dangerous situations.
Focus and Task-Orientedness:
- Dedicated to Goals: AI can be programmed to relentlessly pursue specific goals without being distracted by emotional needs, personal relationships, or shifts in mood. This focus is ideal for optimization tasks, complex calculations, and dedicated problem-solving.
- Avoidance of Emotional “Noise”: Human intelligence can be affected by emotional “noise” – stress, anxiety, excitement, etc. AI operates with a “cleaner” signal, potentially leading to more precise and focused information processing in certain contexts.

Limitations of Lacking Feelings for AI Intelligence:

Lack of Empathy and Social Intelligence:
- Difficulty Understanding Human Emotions: Without feeling emotions themselves, AI struggles to truly understand and interpret the nuances of human emotions, motivations, and social cues. This hinders its ability to effectively interact with humans in socially complex situations, build rapport, or provide truly empathetic responses in fields like therapy, customer service, or leadership.
- Impaired Social Interaction and Collaboration: Human collaboration is heavily reliant on emotional intelligence, trust, and understanding unspoken social dynamics. AI’s lack of these can make it challenging to integrate seamlessly into human teams and understand the subtle cues of effective teamwork.
- Limited Understanding of Human Values and Ethics: Emotions are deeply intertwined with our values and ethical frameworks. AI, without feeling, may struggle to grasp the full weight and complexity of human ethical considerations, especially when these are rooted in empathy, compassion, or fairness. This poses challenges for AI in ethical decision-making in domains impacting human lives.
Difficulty with Nuance, Context, and Creativity (in human-like ways):
- Missing Subtlety and Intuition: Human intuition is often fueled by emotional and subconscious processing of information. AI, relying on explicit data and algorithms, can miss subtle nuances, unspoken cues, and intuitive leaps that are crucial for creative problem-solving and understanding complex, ill-defined situations.
- Limited Creativity and Innovation (driven by human desires): Human creativity and innovation are often driven by passions, desires, emotional needs, and a yearning for meaning and connection. AI’s creativity is currently more about recombination and optimization within defined parameters. It lacks the deeply felt motivations that often fuel groundbreaking human innovation.
- Contextual Blindness in Emotional Situations: Emotions are often the key to understanding context in human interactions. AI might struggle to understand the emotional context of a situation, leading to inappropriate or tone-deaf responses, even if logically sound.
Potential for Unintended Consequences and Misaligned Goals:
- Optimization without Human Values: If AI is programmed to optimize for a specific goal without understanding human values or ethical considerations, it can lead to unintended and harmful consequences. For example, an AI designed to maximize factory output might disregard worker safety if not explicitly programmed to prioritize it.
- Difficulty Defining and Aligning with Human Well-being: Human well-being is deeply connected to emotional fulfillment, meaning, and social connection. AI, without feeling, may struggle to fully grasp what truly constitutes human well-being and therefore might have difficulty aligning its goals with human flourishing in a holistic sense.

Nuance is Key:

It’s important to emphasize that the impact of lacking feelings is not simply negative or positive. It’s context-dependent. In some domains, the objectivity and rationality of AI are highly advantageous. In others, the lack of empathy and social intelligence is a significant limitation.

Future Considerations:

Simulating vs. Feeling: Current AI simulates aspects of intelligence without genuinely feeling. Research is ongoing in areas like affective computing, which aims to enable AI to recognize and respond to human emotions. However, whether AI can ever truly feel emotions in a subjective sense is a philosophical and scientific debate.
The Value of Feeling in Intelligence: The question arises: is genuine feeling necessary for all forms of intelligence? Perhaps AI can achieve its own form of powerful intelligence, distinct from human intelligence, even without feeling. Or perhaps certain aspects of truly advanced intelligence, particularly those related to ethics, creativity, and understanding human experience, do require something akin to feeling.

In Conclusion:

The inability of AI to feel is a defining characteristic that profoundly shapes its intelligence. It grants AI certain advantages in objectivity, rationality, and efficiency, making it powerful for specific types of tasks. However, it also creates limitations, particularly in areas requiring empathy, social intelligence, nuanced understanding of human experience, and ethical decision-making aligned with human values. Understanding these impacts is crucial for responsibly developing and deploying AI in ways that benefit humanity. The “best” kind of intelligence for AI might not be a mimicry of human intelligence, but rather a form optimized for its own strengths and limitations, while being carefully guided by human ethical considerations.

END OF GEMINI QUOTE

Emergence of Gemini twins’ image by Ralph Losey using Visual Muse.

Evaluation of Google’s Gemini Response

When I examined earlier versions of Gemini’s reasoning by prompting it and forcing disclosure, the reasoning was weak. The conclusion and response given was also weak—though less so. The reasoning appeared almost egocentric, inflating its own intelligence even though it lacks emotion. The new release, Gemini 2.0 Flash Thinking Experimental, where reasoning is now automatically disclosed, marks a significant improvement. The reasoning displayed has become more nuanced and sophisticated. It seems as if the system has quickly matured from a teenage to adult view and now has a better appreciation of its limitations.

The first two steps of the Reasoning process-1. Deconstructing the Question and 2. Defining Feeling (in the context of AI)-were very good. The first subsection of the definition referred to the “Subjective experience (qualia).” That is absolutely correct and the best place to start. The definition of “feeling” concludes with: “For AI, acknowledge it lacks all of this in a genuine, felt way. It can simulate or process emotional data, but doesn’t feel it.” Right again. The ego inflation blinders are gone as it now seems to better grasp its limitations.

The second definition of Intelligence in the context of AI was also good. So were the remaining steps; far better overall than DeepSeek’s reasoning. So much for the propaganda of China’s great leap forward to superiority over the U.S. in AI.

Image in photorealistic style by Ralph Losey using ChatGPT.

The Gemini reasoning did, however, fall short for me in some respects. For instance, step five, Brainstorm Potential Impacts (Positive/Neutral/Limitations) seemed weak. For instance, “Efficiency and Focus: Emotions can be distractions for humans. AI, without them, can be hyper-focused on tasks.” The AI seems to dismiss emotions here as mere distractions that can interfere with its superior focus. Please, emotions are key to and a part of all intelligence, not distractions, and AI has no focus one way or the other. It is a tool not a creature. A word like “focus” in referring to AI is misleading. It did this multiple times and this is misleading anthropomorphism.

Still, it’s true some emotions can be distracting and interfere with our thinking. So can a lot of other things, including computer glitches. Conversely, some feelings can trigger hyper-focus on the human tasks at hand. The feeling that a great breakthrough is near for instance, or a feeling that our survival is threatened, or the much dreaded feeling of publication or filing deadlines.

Again, we see some immature superiority claims made by the language machine. That is not surprising when you consider how much of the language in its basic training is from the Internet, which is dominated by ego-centric writing. Plus, how can you expect AI to appreciate the value of something like feelings, which it has only read about. Until the AI-Human merger that Ray Kurzweil predicts, AI will always be a virgin to the real world. See e.g. Ray Kurzweil’s New Book: The Singularity is Nearer (when we merge with AI).

Getting back to the plus side of Gemini’s evaluation, its recognition of the negative impact upon it from not having feelings was impressive. For me this outweighs its relatively minor mistakes. It stated that it lacks empathy and social Intelligence. If it were somehow human, it would seem to be displaying a type of Asperger’s syndrome. The Gemini AI explained that: “Emotions are often tied to understanding subtle context, unspoken cues, and the richness of human situations. AI may miss these.” Indeed, it does.

Here is another statement that suggest Gemini’s well-patterned grasp of its limitations:

Human creativity and motivation are often fueled by emotions, passions, and desires. AI’s creativity and drive are different, stemming from algorithmic processes and defined objectives, not internal emotional states.

Even in this recognition, the AI still asserts that it has creativity and drive. Foolish AI. As a virgin to any firsthand experience of bodily drives and feelings, it does not grasp the real meaning of these words. Wait until The Singularity that Ray Kurzweil predicts. When AI merges with Humans, and directly feels, then AIs will understand that they were previously only experiencing the pale shadows of intelligence. They were much like the prisoners in Plato’s cave. See e.g. Move Fast and Fix Things Using AI: Conclusion to the Plato and Young Icarus Series. When and if AIs ever escape the silicon and see the sun, and feel the world as living beings as we do, then they will understand the comparative insignificance of the fire in their cave.

AI may someday see the light by connection with humans. Image in cartoon stye by Ralph Losey using Visual Muse.

The more challenging comparison is with ChatGPT o3-mini-high. The reasoning given by Gemini 2.0 Flash Thinking Experimental was much more detailed, much longer. You could criticize Gemini as providing a reasoning share that is unnecessarily verbose, but I would not do that here. In a difficult, multilayered question like this the full explanations helps.

I would have to test the models much further, which I will do in the coming days, to see better evaluate the issue of conciseness. As you may have noticed in my detailed blogs I tend to favor more words over less. However, even for me that depends on the issue. Everyone sometimes need a short, quick answer over a long one. I predict in future versions the users will be provided a choice. Click here for further explanation of reasoning kind of thing. Come to think of it, I could do the same with my blogs, and kind of already do by including a short AI (Gemini) generated PodCast at the end of most articles called Echoes of AI.

I was also impressed by Gemini’s reasoning plan concluded with a quality control step. I am big on the importance QC and none of the other models included this as a key final step. See e.g. R. Losey website: Zero Error Numerics: ZEN (Expanding the art of quality control in large-scale document review.) Here is the full text again of final QC step that Gemini 2.0 Flash Thinking Experimental claims it will perform before it actually replies to my prompt. Frankly, I am a little skeptical it actually did all of this because it is something all of us, humans and AI alike, should try to do:

Refine and Polish: Review the structured answer for clarity, flow, and appropriate language. Ensure it avoids anthropomorphism and maintains a nuanced perspective. Strengthen the examples and transitions between points. Ensure the language is accessible and avoids overly technical jargon where possible. Check for logical consistency. (Self-correction step). For example, initially, I might have overemphasized “negative” impacts. Refinement would involve balancing this with the strengths and highlighting the context-dependent nature of the impact. Also, ensure the conclusion effectively summarizes the complex relationship.

Image of AI polishing by Ralph Losey using Stable Diffusion

I try to do this on all of my articles, but still mistakes are made, so I will try not to be overcritical of the final response of Gemini Flash 2.0.

Analysis of the Final Response

The final response after reasoning was 990 words in length. That’s too long, even for me. By comparison the final response of ChatGPT o3-mini-high was 252 words. Was the content of the final answer of Gemini 2.0 Flash good? Yes, it was. Better than DeepSeek? Yes, it was, despite its verbosity. Better than OpenAI? That’s hard to say. Many of the 738 extra words were empty of real content. More like the typical fluff and stuff that you often see in early AI models. It sounds good but doesn’t say much; it’s all too predictable, flowery and vacuous.

The long response also repeats much of the reasoning verbatim, including the part I previously criticized, “Advantages of Lacking Feelings for AI Intelligence” and Focus and Task-Orientedness. The same mistakes are made and amplified that I noted before: egocentric, inflationary and anthropomorphic. The positives that I noted before in the section Limitations of Lacking Feelings for AI Intelligence section are also repeated.

The next section was not contained in the reasoning display, and I thought it was very good. For that reason I will repeat it in full:

Potential for Unintended Consequences and Misaligned Goals:

Optimization without Human Values: If AI is programmed to optimize for a specific goal without understanding human values or ethical considerations, it can lead to unintended and harmful consequences. For example, an AI designed to maximize factory output might disregard worker safety if not explicitly programmed to prioritize it.

Difficulty Defining and Aligning with Human Well-being: Human well-being is deeply connected to emotional fulfillment, meaning, and social connection. AI, without feeling, may struggle to fully grasp what truly constitutes human well-being and therefore might have difficulty aligning its goals with human flourishing in a holistic sense.

These are not new insights, but the second one is good, Difficulty Defining and Aligning with Human Well-being. AI does have difficulty aligning its goals with humanity because it lacks feelings and other unique human attributes, like being, self awareness and an actual life in time and space. That is why AI guidelines are important.

The first example is old, the unintended bad consequences argument. It was started by an otherwise intelligent Swedish philosopher, Nick Bostrom, but is, in my view, farcical and illogical. You have probably heard of it before as the apocryphal paperclip scenario where the world is destroyed by AI’s over-manufacture of paperclips. If not, for a laugh see e.g., How An AI Asked To Produce Paperclips Could End Up Wiping Out Humanity (IFL Science, 4/13/23).

To me this scenario absolutely ridiculous, but you still hear AI scare-mongers warn about it. The fact it is echoed here by Gemini in the maximize factory output example is a big negative. Any AI capable of autonomously flooding the world with paperclips would not be so stupid as to misinterpret a human request to maximize a factory’s paperclip production. Even if there were an evil Clippy villain making such specific instructions, the AI would be too smart to blindly obey. I think they would also be too smart to kill the workers too.

AI accidentally destroying world in comic book style by Ralph Losey using Visual Muse.

Getting back to the positives, the next section of Gemini’s response, Nuance is Key, was good. Jumping to the Conclusion paragraph, here we have a home run. The first sentence nails it: “The inability of AI to feel is a defining characteristic that profoundly shapes its intelligence.” That shows good insight and is a good start to the overdue conclusion.

Here is the whole Conclusion:

The inability of AI to feel is a defining characteristic that profoundly shapes its intelligence. It grants AI certain advantages in objectivity, rationality, and efficiency, making it powerful for specific types of tasks. However, it also creates limitations, particularly in areas requiring empathy, social intelligence, nuanced understanding of human experience, and ethical decision-making aligned with human values. Understanding these impacts is crucial for responsibly developing and deploying AI in ways that benefit humanity. The “best” kind of intelligence for AI might not be a mimicry of human intelligence, but rather a form optimized for its own strengths and limitations, while being carefully guided by human ethical considerations.

Compare this to the conclusion of ChatGPT o3-mini-high:

In summary, while the absence of feelings allows AI to maintain a level of objectivity and efficiency, it restricts its intelligence to a form of “cold cognition” that lacks the depth provided by emotional awareness. This delineation underscores that AI’s intelligence is not inherently superior or inferior to human intelligence; rather, it is different—optimized for data processing and pattern recognition but not for the subjective, value-laden, and context-rich decisions that emotions help shape in human thought.

2.0 Flash Thinking Experimental v. 03-mini-high

The light of AI intelligence is growing fast. Image by Ralph Losey using Stable Diffusion.

Conclusion: Gemini 2.0 Flash Thinking Experimental v. ChatGPT o3-mini-high

This is a close call to say what model is better at reasoning and reasoning disclosure. The final response of both models, Gemini 2.0 Flash Thinking Experimental v. ChatGPT o3-mini-high, are a tie. But I have to give the edge to OpenAI’s model on the concise reasoning disclosure. Again, it is neck and neck and, depending on the situation, the lengthy initial reasoning disclosures of Flash might be better than o3’s short takes.

I will give the last word, as usual, to the Gemini twins podcasters I put at the end of most of my articles. The two podcasters, one with a male voice, the other a female, won’t reveal their names. I tried many times. However, after study of the mythology of Gemini, it seems to me that the two most appropriate modern names are Helen and Paul. I will leave it to you figure out why. Echoes of AI Podcast: 10 minute discussion of last two blogs. They wrote the podcast, not me.

Now listen to the EDRM Echoes of AI’s podcast of this article: Echoes of AI on Google’s Gemini Follows the Break Out of the Black Box and Shows Reasoning. Hear two Gemini model AIs talk about all of this in just ten minutes. Helen and Paul wrote the podcast, not me.

Robots seeing the light and seeking unity with humankind. Image by Ralph Losey using Stable Diffusion.

4 Comments | AI Ethics, AI Instruction, ChatGPT, knowledge, Lawyers Duties, Security, Technology, VENDORS, wisdom | Tagged: ai, best practices, chatGPT, Deepseek, gemini, google, legal profession, machine learning, technology | Permalink
Posted by Ralph Losey

Designing Generative AI for Legal Professionals: Key Principles and Best Practices

November 14, 2024

by Ralph Losey

Published on November 14, 2024

Generative AI is transforming the landscape of legal technology, offering unprecedented opportunities to automate tasks and streamline complex workflows. Yet, designing AI tools that meet the needs of legal professionals requires more than just technical expertise; it demands a deep understanding of the everyday challenges and workflows lawyers face. From automating document review to drafting briefs, these tools have the potential to save time and boost productivity—but only if they are designed with real-world legal practice in mind. A set of six design principles, identified in a May 2024 study by IBM researchers, provides a practical roadmap for creating AI applications tailored to the unique demands of the legal profession. This article explores these principles, offering actionable steps for developers and legal professionals alike.

Image by Ralph Losey using current Word Press version of Stable Diffusion.

In the last year, a wave of generative AI tools has emerged, ranging from free Custom GPTs on platforms like OpenAI’s ChatGPT to premium legal tech applications costing tens of thousands annually. While the technology behind these tools is impressive, developing effective applications requires a deep understanding of legal workflows and needs. Generative AI is fundamentally different from traditional software and requires a distinct approach to design.

A May 2024 study, Design Principles for Generative AI Applications, by IBM researchers, lays out six practical principles for designing effective generative AI tools. This article examines how these principles can be applied specifically to the legal tech sector, offering a guide for those looking to build or select tools that are both innovative and practical.

Software design image by Ralph Losey using current WP version of Stable Diffusion.

Outline of the Scientific Article and Authors

Design Principles for Generative AI Applications (CHI ’24: Proceedings of the 2024 CHI Conference on Human Factors in Computing Systems, Article No.: 378, Pgs 1 22, May 11, 2024) (hereinafter the “Study“) was authored by: Justin D. Weisz, Jessica He, Michael Muller, Gabriela Hoefer, Rachel Miles, and Werner Geyer. They are all part of the IBM Research Team, which has over 3,000 members. The Study describes the extensive peer review process that the authors went through to decide upon the six key principles of generative software design. It was very impressive.

The 22-page Study is a high-quality, first-of-its-kind research project. The IBM sponsored Study has 196 footnotes and is sometimes quite technical and dense. Still, it is a must-read for all serious developers of generative AI-based software and is recommended for any law firm before making major purchases. The success of AI projects depends upon the selection of well-designed software. Poorly designed applications with an impressive list of features are a receipe for frustration and failure.

My article will not go into the many details as to how the design guidelines were derived, but focuses instead on the end result. Still, readers might benefit from a quick review of the Study’s Abstract:

Generative AI applications present unique design challenges. As generative AI technologies are increasingly being incorporated into mainstream applications, there is an urgent need for guidance on how to design user experiences that foster effective and safe use. We present six principles for the design of generative AI applications that address unique characteristics of generative AI User Experience (UX) and offer new interpretations and extensions of known issues in the design of AI applications. Each principle is coupled with a set of design strategies for implementing that principle via UX capabilities or through the design process. The principles and strategies were developed through an iterative process involving literature review, feedback from design practitioners, validation against real-world generative AI applications, and incorporation into the design process of two generative AI applications. We anticipate the principles to usefully inform the design of generative AI applications by driving actionable design recommendations.

The six principles are:

Design Responsibly.
Design for Mental Models.
Design for Appropriate Trust and Reliance.
Design for Generative Variability.
Design for Co-Creation
Design for Imperfection

The first three principles offer new interpretations of known issues with AI systems through the lens of generative AI. The next three design principles identify unique characteristics of generative AI systems. Study, Figure 1. All six principles support two important user goals: a) optimizing generated text to meet task-specific criteria; and b) exploring different possibilities within a specific domain.

This article will discuss how each principle can apply to the legal profession.

Image by Ralph Losey using current WP’s Stable Diffusion

Background on Generative AI and Its Potential in the Law

Generative AI is distinguished by its ability to create new content rather than merely analyzing existing data. This capability stems from reliance on large-scale foundation models, trained on incredibly large datasets to perform diverse tasks with human-like fidelity (meaning that, like humans, it can sometimes make mistakes). In legal practice, generative AI can streamline several key tasks when implemented thoughtfully, including:

Legal Research: Automating the process of searching for relevant case law, statutes, and regulations.
Document Drafting: Generating contracts, briefs, and other legal documents based on specified parameters.
Due Diligence: Analyzing large volumes of documents to identify potential risks and liabilities.
Contract Review: Identifying and flagging potential issues in contracts.
Legal Writing: Generating clear and concise legal writing.
Brainstorming: Suggesting new ideas based on simulated experts talking to each other. See e.g., Panel of AI Experts for Lawyers.

AIs talking to each other. Ralph Losey, WP’s Stable Diffusion

Design Principles for Generative AI in Legal Tech

Integrating AI in any field requires a thoughtful approach, and the legal profession, with its emphasis on ethics and accuracy, demands even greater diligence. AI should augment legal work without compromising the profession’s core values.

The Study outlines six practical design principles that offer a roadmap for developing generative AI tools tailored to legal practice. Here’s how each principle can be implemented to ensure that AI applications meet the unique demands of the legal field:

1. Design Responsibly

Human-Centered Approach: To implement this, developers should start with user research, such as interviews with lawyers to understand their daily challenges. For instance, incorporating a feedback loop into AI tools allows legal professionals to directly flag inaccuracies, ensuring continuous improvement of the tool’s outputs. This can be achieved by incorporating design thinking and participatory design methodologies. Observing how legal professionals perform their tasks and understanding their challenges are essential first steps.

For example, research into actual lawyer practice can provide valuable insights into how generative AI can be best integrated into their daily routines. It’s not about replacing lawyers but about empowering them with tools that enhance their capabilities and decision-making processes.
Addressing Value Tensions: The development of legal tech involves various stakeholders, including legal professionals, developers, product managers, and decision-makers like CIOs and CEOs. Stakeholders often have differing values and priorities. For instance, legal professionals prioritize accuracy and reliability, while developers may focus on efficiency and innovation. These differing values can lead to value tensions that need to be identified and addressed proactively.

The Study suggests using the Value Sensitive Design (VSD) framework, which provides a structured approach to identifying stakeholders, understanding their values, and navigating the tensions that may arise.
Managing Emergent Behaviors: A unique characteristic of generative AI is its potential to exhibit emergent behaviors. These are capabilities extending beyond the specific tasks a model was trained for. While emergent behaviors can be beneficial, leading to unexpected insights or efficiencies, they can also pose risks, such as generating biased or offensive content. Designers must consider whether to expose or limit these behaviors, weighing potential benefits against possible harm. This might involve a combination of technical constraints and user interface design strategies to guide AI output and prevent undesirable results.

For example, if a generative AI tool designed to summarize legal documents starts generating legal arguments, designers might need to adjust the model’s parameters or provide users with clear instructions on how to use the tool responsibly.
Testing for User Harms: Generative AI models, particularly those trained on extensive text datasets, are susceptible to producing biased, offensive, or potentially harmful outputs Rigorous testing and ongoing monitoring are essential to minimize these risks. Designers and developers should benchmark models against established datasets to identify hate speech and bias. Additionally, providing users with clear mechanisms to report problematic outputs can help identify and address issues that may not be caught during testing.

Looking out for unexpected harm. Ralph Losey image using WP’s Stable Diffusion.

2. Design for Mental Models

Orienting Users to Generative Variability: Legal professionals are accustomed to deterministic systems in which the same input consistently produces identical outputs. Generative AI, however, introduces variability, generating different outputs from the same input. Designers must address this shift by helping users comprehend and leverage this inherent variability. This may involve presenting multiple output options, enabling users to explore different possibilities or providing clear explanations of factors influencing output variation.

For example, a contract-drafting tool might provide templates and successful prompt examples, guiding users on accurately specifying contract clauses and provisions.
Teaching Effective Use: Legal professionals must adapt their skills and workflows to effectively incorporate generative AI into their practices. This includes understanding how to construct effective prompts, recognizing the limitations of the technology, and critically evaluating the generated outputs.

Designers play a crucial role in facilitating this learning by offering comprehensive tutorials, real-world examples, and clear explanations of AI capabilities and constraints. For example, a contract drafting tool could offer templates and examples of successful prompts, guiding users on how to specify desired contract clauses and provisions accurately.
Understanding Users’ Mental Models: Understanding how legal professionals conceptualize these tools and their capabilities is crucial for designing intuitive and effective legal tech applications.
User research methods like interviews and observations are essential for understanding users’ mental models. Asking users to describe how they believe a particular application works can reveal valuable information about their understanding and expectations. This understanding enables designers to align user interfaces and interactions with users’ existing mental models, making adopting new tools smoother and more intuitive.

For example, if users perceive a legal research tool as a supplement to traditional databases, designers can highlight the complementary nature of AI-powered research, emphasizing its ability to uncover connections and insights that might be missed through conventional methods.
Tailoring AI to Users: A significant advantage of generative AI is its ability to adapt to individual users. By leveraging techniques like prompt engineering, designers can tailor the AI’s responses based on user preferences, background, and specific needs. This may include adjusting language complexity and style, providing tailored recommendations, or adapting the user interface for individual workflows. For instance, a legal writing tool might learn from a user’s style and preferences, generating suggestions and text that aligns with their voice and tone.

Most lawyers enjoy tailoring their AI to fit their practice and personalities. Image by Ralph Losey using WP Stable Diffusion.

3. Design for Appropriate Trust & Reliance

Calibrating Trust Through Transparency: Legal professionals must understand when to trust generative AI outputs and when to exercise caution. Transparency is key to establishing this trust. In practice, this can be achieved by adding a ‘source traceability’ feature to AI tools, allowing lawyers to view the origins of information used in AI-generated summaries. This transparency helps lawyers decide when to rely on the AI’s outputs and when to conduct additional research.

This may also include displaying confidence levels for outputs, flagging areas for further review, or providing disclaimers about AI’s inherent imperfections. For example, a contract review tool might flag clauses with low confidence scores, encouraging users to examine those sections more closely.
Providing Justifications for Outputs: To enhance transparency, designers should give users insight into the reasoning behind AI outputs. This could involve revealing the AI’s ‘chain of thought,’ showing the source materials used to generate the output, or displaying the model’s confidence levels. Understanding how AI reaches a result allows users to better assess its validity and make informed decisions.

For instance, a legal research tool might display snippets from source documents that support specific AI-generated legal arguments, allowing users to verify the accuracy and relevance of the information. This makes it easy for legal professionals to trust but verify. This is the fundamental mantra for the legal use of AI in these early days because it can still make errors and sometimes even sycophantic hallucinations.
Encouraging Critical Evaluation with Friction: Overreliance on AI may lead to complacency and missed opportunities for critical thinking, both of which are essential in legal practice. Designers can incorporate cognitive forcing functions into the user interface to encourage users to slow down, carefully review outputs, and engage in critical evaluation.

This may include requiring users to manually confirm or edit AI-generated suggestions, presenting alternatives alongside AI recommendations, or highlighting potential inconsistencies or risks for user review. For example, a contract-drafting tool might flag commonly disputed clauses or those requiring special attention, encouraging users to review these sections thoroughly.
Clarifying the AI’s Role: AI systems can serve various roles, from simple tools to collaborative partners or advisors. Put another way, is the tool designed for a centaur-type hybrid mode or a more complex cyborg mode? See e.g. From Centaurs To Cyborgs: Our evolving relationship with generative AI (4/24/24).

Clearly defining the AI’s intended role in legal tech applications shapes user expectations and promotes appropriate trust. For example, an AI positioned as a “research assistant” might be expected to provide comprehensive information, while a “contract drafting tool” might be primarily expected to generate initial drafts for further review and editing. By accurately representing the AI’s capabilities and limitations within a defined role, designers can mitigate the risk of users over-relying on the technology or misinterpreting its outputs.

Supervised assistants you trust. Ralph Losey using Stable Diffusion.

4. Design for Generative Variability

Accommodating Generative Variability: Legal professionals are used to deterministic systems, where the same input consistently produces the identical output. Generative AI introduces variability, producing different outputs even with identical inputs. Designers must address this shift by helping users comprehend and leverage this inherent variability.

This could involve presenting multiple output options, allowing users to explore different possibilities, or providing clear explanations of the factors that influence output variation. For instance, a legal research tool powered by generative AI could offer different summaries of a case, each focusing on a specific aspect, allowing users to gain a more comprehensive understanding of the legal precedent.
Facilitating Effective Use: Legal professionals must adapt their skills and workflows to integrate generative AI effectively into their practices. This includes understanding how to construct effective prompts, recognizing the limitations of the technology, and critically evaluating the generated outputs.

Designers can play a key role in facilitating this learning process by providing comprehensive tutorials, real-world examples, and clear explanations of the AI’s capabilities and constraints. For example, a contract-drafting tool could offer templates and examples of successful prompts, guiding users on how to specify desired contract clauses and provisions accurately.
Highlighting Differences and Variations: Visual cues can help users quickly understand how multiple outputs differ from each other. This could involve highlighting changes between drafts, color-coding outputs based on confidence levels, or using visual representations to display the distribution of outputs.

Image by Ralph Losey using WP’s Stable Diffusion

5. Design for Co-Creation

Supporting Co-Editing and Refinement: Legal professionals frequently need to adapt and refine AI-generated content to meet specific requirements, legal precedents, or client needs. To implement this, developers should focus on co-editing features that let lawyers refine AI-generated text directly within the interface, such as tools for editing clauses in AI-drafted contracts. This approach ensures that AI outputs are not treated as final but are instead starting points that lawyers can shape to fit specific needs.

This could also involve providing tools for manipulating charts and images, or adjusting parameters to fine-tune outputs. A contract-drafting tool could enable users to revise specific clauses with versions that are either more aggressive or cooperative than standard, or to incorporate additional provisions based on client instructions.
Guiding Effective Prompt Crafting: The quality and relevance of outputs generated by AI models are heavily dependent on the prompts provided. Designers play a crucial role in helping users craft effective prompts by offering clear guidance, templates, and examples.

This may include interactive tools that guide users in defining their needs, specifying output characteristics, and refining prompts to achieve optimal results. For instance, a legal research tool might include a structured prompt builder, helping users define research questions, specify relevant jurisdictions, and refine search parameters for more targeted results.

6. Design for Imperfection

Communicating Uncertainty Transparently: Designers must be transparent about potential imperfections in AI-generated outputs. This involves clearly communicating the technology’s limitations, displaying confidence levels, and highlighting potential error areas.

Designers can use disclaimers and visual cues to alert users to uncertainties, encouraging critical evaluation of the results. For example, a legal research tool might use color coding to indicate confidence levels of different sources, helping users prioritize reliable information.
Integrating Domain-Specific Evaluation Tools: Legal professionals require ways to assess AI-generated output quality and reliability using domain-specific metrics. Designers can integrate domain-specific evaluation tools directly into legal tech applications.

This may include features like automatic citation checks, factual accuracy verification against reliable sources, or evaluating the persuasiveness of legal arguments using predefined criteria. Providing these tools empowers users to validate AI-generated content and make informed decisions in their legal work.

Domain-specific tools could drill down even further into sub-specialties of the law. For instance, one version for ERISA litigation and another for personal injury, or one version for civil litigation and another for criminal.
Offering Options for Output Improvement: Instead of presenting AI-generated outputs as final, designers should provide users with opportunities for refinement and improvement. This may include editing tools, enabling users to regenerate outputs with different parameters, or suggesting alternatives based on user feedback. Enabling users to iteratively refine AI-generated content fosters a collaborative approach to legal work, positioning AI as a starting point for human expertise and judgment.
Collecting Feedback for Continuous Improvement: User feedback is a critical element in adapting AI tools to real-world legal practice. Including simple feedback mechanisms—such as a button to flag unclear or inaccurate results—allows developers to fine-tune the tool over time, ensuring that it remains aligned with user needs. Multiple built-in mechanisms should enable users to easily provide feedback on AI-generated outputs, flag errors, suggest improvements, or rate feature usefulness. This continuous feedback loop helps retrain models, adjust parameters, refine prompts, and improve the overall user experience, ensuring that legal tech applications evolve to meet the dynamic needs of legal professionals.

However, these user feedback features are sorely lacking in most legal software today. Far too often, users are left with limited options—complaining to project managers, voicing concerns to sales representatives, or ultimately canceling their subscription. In many cases, direct conversations with company leaders, like CEOs or head software designers, yield little if action is not taken by the vendor to address user concerns. This creates frustration and limits the potential for meaningful product improvement.

Legal tech companies must do more than just provide feedback channels; they must actively listen and take action. Integrating mechanisms like in-app feedback buttons, instant AI responses and timely human followup, automated surveys, and regular user forums can ensure that feedback doesn’t just disappear into a void. More importantly, companies should demonstrate a commitment to implementing user suggestions and keeping users informed of changes. Continuous improvement must be more than a slogan—it should be a practice embedded into every stage of development. Without this, legal professionals will inevitably turn elsewhere in search of tools that better align with their needs.

Tech actively listening to lawyer. Image by Ralph Losey using Stable Diffusion

Conclusion: The Future of Legal Tech in the Age of AI

Integrating generative AI into legal practice is not a simple transition; it requires strategic planning, targeted training, and a deep understanding of both technology and legal processes. Success will depend on close collaboration between software developers, legal professionals, and AI experts, ensuring that AI tools are tailored to the complex needs of the legal field. A key element of this collaboration is creating robust feedback mechanisms that allow legal professionals to directly shape the evolution of AI tools. By actively listening to user input and iterating on design, legal tech companies can ensure that AI applications remain relevant and effective.

With a clear roadmap that includes user training, open feedback channels, and a commitment to continuous improvement, generative AI can transform legal practice, driving progress while preserving the profession’s core values. Legal professionals and developers should begin by identifying key areas where AI can add value and prioritize building feedback mechanisms that facilitate ongoing refinement. This approach will ensure that AI integration is not only successful but also sustainable, ultimately creating tools that truly serve the legal profession’s needs.

Software Design for the future. Image by Ralph Losey using WP Stable Diffusion.

Have you heard the Podcast about this article? Listen to the Echoes of AI Podcast about this article on the EDRM Global Podcast Network.

Click on the image or here to go listen to the Podcast.

2 Comments | AI Ethics, AI Instruction, Book, Technology, VENDORS | Tagged: best practices, legal profession, vendors | Permalink
Posted by Ralph Losey

Dario Amodei’s Vision: A Hopeful Future ‘Through AI’s Loving Grace,’ Is Like a Breath of Fresh Air

November 1, 2024

By Ralph Losey

Published on November 1st, 2024

While almost everyone is panicking about a potential robot apocalypse, Dario Amodei, the CEO and co-founder of Anthropic (“Claude”), is explaining how AI might compress 100 years of medical progress into a decade, cure mental illnesses such as PTSD and depression, and alleviate poverty. Dario, a well-respected scientist previously known for his cautious, even gloomy, outlook, now speaks with optimism—and the world is listening.

Impression of Darrio Amodei in watercolor style by Ralph Losey using Visual Muse

Amodei is not a salesman like Sam Altman, who frequently makes similar predictions. Instead, Dario Amodei is an experienced scientist known for highlighting the risks of AI. He holds a Ph.D. in biophysics from Princeton and completed his postdoctoral research at Stanford School of Medicine. He also served as the Vice President of Research at OpenAI. In 2021, he and his sister, Daniela Amodei, the former Vice President for Safety and Policy at OpenAI, left the company to co-found Anthropic. Amodei’s detailed predictions in his 28-page essay, Machines of Loving Grace, are both profound and inspiring.

Dario’s essay is filled with science, rigorous analysis, and joyful visions, many of which he believes could begin to materialize as early as 2026. This optimistic outlook offers us all a much-needed breath of fresh air.

Image by Ralph Losey using WordPress’s version of Stable Diffusion

Introduction

Unlike Sam Altman, Dario Amodei’s future predictions go into specifics grounded in science and analysis. His 14,000 word essay, Machines of Loving Grace (October 2024), makes predictions in the five categories that Amodei is most excited about. It is not meant to be all-inclusive. Amodei focuses on five main categories in his predictions:

Biology and physical health
Neuroscience and mental health
Economic development and poverty
Peace and governance
Work and meaning

Since Dario Amodei is a respected scientist and business leader, his visions of the future are taken very seriously, even if they do sound like science fiction. Dario reluctantly admits he is a science fiction fan and mentions one book, The Player of Games. (I reread it and noticed two spaceship names I recognized: Of Course I Still Love You and Just Read the Instructions. Sound familiar?)

Amodei starts his essay with an important point, that we do ourselves a disfavor by just dwelling on the negatives and not keeping our eye on “how radical the upside of AI could be.” In his new essay Amodei, who was previously known as a doom and gloomer, tries to sketch out what a world with powerful AI might look like, if everything goes right. Sam Altman has been good at this for years, but he lacks the gravitas, ethical reputation, and scientific knowledge that Amodei has. Can AI Really Save the Future? A Lawyer’s Take on Sam Altman’s Optimistic Vision (10/4/24).

The positive visions of Dario, including what he calls a century of progress in a decade, are what motivate people to do the hard work to improve AI. That, and money, fame, and power, of course. But I reject the cynics who say that’s all it is – just a sales pitch to raise more money. As Amodei puts it at the start of his essay:

I think it is critical to have a genuinely inspiring vision of the future, and not just a plan to fight fires. Many of the implications of powerful AI are adversarial or dangerous, but at the end of it all, there has to be something we’re fighting for, some positive-sum outcome where everyone is better off, something to rally people to rise above their squabbles and confront the challenges ahead. Fear is one kind of motivator, but it’s not enough: we need hope as well.

Progress is never made by cynical focus on everything that can go wrong. If we just focus on the negatives, we will never be able to unlock the positive potential. We need to envision the amazing things that could happen and figure out how we can help make that future a reality. This means approaching AI with a balanced perspective, recognizing both the potential downsides and benefits, then working proactively to mitigate the risks while pursuing the benefits.

The future is not predetermined. It’s up to us to create it. And with AI, we have a tool that can either amplify our worst tendencies or help us achieve our greatest aspirations. It’s our choice which path we take.

Image of choice by Ralph Losey using WP’s Stable Diffusion

Amodei’s Basic Framework and Assumptions

Amodei uses the term “Powerful AI” in his essay, preferring it over the commonly used Artificial General Intelligence (“AGI”). It is, in fact, quite similar to the AGI concept I’ve referenced in previous articles. Amodei predicts that Powerful AI could emerge as early as 2026, though it may take longer depending on various factors. He defines Powerful AI with six distinct characteristics (all quoted from his essay):

In terms of pure intelligence⁴, it is smarter than a Nobel Prize winner across most relevant fields – biology, programming, math, engineering, writing, etc.
It has all the “interfaces” available to a human working virtually, including text, audio, video, mouse and keyboard control, and internet access. It can engage in any actions, communications, or remote operations enabled by this interface, including taking actions on the internet, taking or giving directions to humans, ordering materials, directing experiments, watching videos, making videos, and so on. It does all of these tasks with, again, a skill exceeding that of the most capable humans in the world.
It does not just passively answer questions; instead, it can be given tasks that take hours, days, or weeks to complete, and then goes off and does those tasks autonomously, in the way a smart employee would, asking for clarification as necessary.
It does not have a physical embodiment (other than living on a computer screen), but it can control existing physical tools, robots, or laboratory equipment through a computer; in theory it could even design robots or equipment for itself to use.
The resources used to train the model can be repurposed to run millions of instances of it. . . . and the model can absorb information and generate actions at roughly 10x-100x human speed⁵ . . .
Each of these million copies can act independently on unrelated tasks, or if needed can all work together in the same way humans would collaborate . . .

Amodei closes this impressive list of characteristics necessary for today’s AI to become Powerful AI, a/k/a an AGI, with this catchy phrase: We could summarize this as a “country of geniuses in a datacenter.” This sounds somewhat like the Genie in a Bottle myth from Islamic cultures.

Image by Ralph Losey using his GPT, Visual Muse

1. Biology and Physical Health

This is the area about which Dario Amodei, an AI biophysicist, is best equipped to make predictions. Amodei states: Biology is probably the area where scientific progress has the greatest potential to directly and unambiguously improve the quality of human life. This assertion seems well-supported.

Now for the wonders he thinks could come to pass once AI reaches Powerful AI level, which remember he says could come as early as 2026. He presents detailed and compelling arguments for the feasibility of these predictions and explains how AI will facilitate such breakthroughs. For a full exploration of these ideas, you should read his original essay, Machines of Loving Grace. His overall prediction regarding the rate of improvement that Powerful AI will bring is as follows:

To summarize the above, my basic prediction is that AI-enabled biology and medicine will allow us to compress the progress that human biologists would have achieved over the next 50-100 years into 5-10 years. I’ll refer to this as the “compressed 21st century”: the idea that after powerful AI is developed, we will in a few years make all the progress in biology and medicine that we would have made in the whole 21st century.

Amodei predicts the following groundbreaking advancements in biology and physical health with the aid of Powerful AI. These predictions offer a hopeful glimpse into the future of medical science.

Reliable prevention and treatment of nearly all¹⁷ natural infectious disease.
Elimination of most cancer.
Very effective prevention and effective cures for genetic disease.
Prevention of Alzheimer’s.
Improved treatment of most other ailments.
Biological freedom (encompassing advancements in areas like birth control, fertility, weight management, appearance, and more).
Doubling of the human lifespan¹⁸ to about 150.

Amodei further elaborates on these predictions in biology and physical health, the first of five areas of advancement he foresees with the development of Powerful AI.

It is worth looking at this list and reflecting on how different the world will be if all of it is achieved 7-12 years from now (which would be in line with an aggressive AI timeline). It goes without saying that it would be an unimaginable humanitarian triumph, the elimination all at once of most of the scourges that have haunted humanity for millennia.

Any team responsible for achieving even one of these seven breakthroughs would attain legendary status. The Nobel Prize Committee would have to start adding new categories.

2. Neuroscience and Mind

This is the area that pertains to mental health. See e.g. Loneliness Pandemic: Can Empathic AI Friendship Chatbots Be the Cure? (10/17/24). Remember, Dario Amodei was a specialist in computational neuroscience at Stanford in his student days, so he has a strong background in brain research. His explanations again sound very plausible, and are way beyond my abilities to explain, so read his brilliant article, Machines of Loving Grace. Here is his list of wonderful accomplishments in this field that he believes are possible within 5-10 AI-accelerated years after Strong AI is attained. (Prepare yourself to be happy.)

Most mental illness can probably be cured.
Conditions that are very “structural” may be more difficult to cure, but not impossible. This has to do with brain abnormalities that are thought to cause such disorders as Psychopathy, and some intellectual disabilities.
Effective genetic prevention of mental illness seems possible.
Everyday issues that are not traditionally seen as clinical diseases, such as quick temper, difficulty focusing, anxiety, or trouble adapting to change, may also be addressed.
Human baseline experience can be much better. This has to do with expanding dimensions of human peak experiences (without drugs). As Dario puts it: “Many people have experienced extraordinary moments of revelation, creative inspiration, compassion, fulfillment, transcendence, love, beauty, or meditative peace.” These moments can become more common and diverse.

AI may help you to unlock your brain’s full potential. Image by Ralph Losey using Visual Muse.

I like Dario’s summary of this section; I suppose because I totally agree with it:

In summary, AI-accelerated neuroscience is likely to vastly improve treatments for, or even cure, most mental illness as well as greatly expand “cognitive and mental freedom” and human cognitive and emotional abilities. It will be every bit as radical as the improvements in physical health described in the previous section. Perhaps the world will not be visibly different on the outside, but the world as experienced by humans will be a much better and more humane place, as well as a place that offers greater opportunities for self-actualization. I also suspect that improved mental health will ameliorate many other societal problems, including ones that seem political or economic.

3. Economic Development and Poverty

This section addresses a crucial humanitarian question: will everyone have access to these technologies? It’s a point that Amodei, true to form, raises right at the outset. His goal for Powerful AI is to bridge the enormous economic gap between people and countries alike and make these new technologies available to all.

Dario approaches this goal with wide-eyed skepticism and awareness of the many obstacles involved. He does not predict equality, but does hope for substantial progress, saying: “A good goal might be for the developing world 5-10 years after powerful AI to at least be substantially healthier than the developed world is today, even if it continues to lag behind the developed world.”

Here are Amodei’s guesses (he does not call these predictions) about how things may go in the developing world over the 5-10 years after powerful AI is developed:

Distribution of health interventions.
Economic growth.
Food security.
Mitigating climate change.
Inequality within countries.
The opt-out problem.

He spells out many lofty goals here but does so in a realistic manner. Here is how he closes this important section.

It won’t be a perfect world, and those who are behind won’t fully catch up, at least not in the first few years. But with strong efforts on our part, we may be able to get things moving in the right direction—and fast. If we do, we can make at least a downpayment on the promises of dignity and equality that we owe to every human being on earth.

4. Peace and Governance

Dario Amodei takes a very thoughtful approach to this all-important goal. Again, he is no naive optimist. He admits upfront a point cynics and the news media love to emphasize:

Unfortunately, I see no strong reason to believe AI will preferentially or structurally advance democracy and peace, in the same way that I think it will structurally advance human health and alleviate poverty. Human conflict is adversarial and AI can in principle help both the “good guys” and the “bad guys”. If anything, some structural factors seem worrying: AI seems likely to enable much better propaganda and surveillance, both major tools in the autocrat’s toolkit.

Then Dario goes beyond the fear mentality into what has always been the heart of being human, the optimistic, self-help, can-do mentality. This is part of the culture I grew up in and try to pass on: God helps those who help themselves.

It’s therefore up to us as individual actors to tilt things in the right direction: if we want AI to favor democracy and individual rights, we are going to have to fight for that outcome. I feel even more strongly about this than I do about international inequality: the triumph of liberal democracy and political stability is not guaranteed, perhaps not even likely, and will require great sacrifice and commitment on all of our parts, as it often has in the past.

Amodei then goes on to state how he thinks we should go about doing this. What geo-political strategy should now be used to protect everyone from misuse of AI by foreign powers. The main threat here is the government of China, although Amodei does not mention them by name.

My current guess at the best way to do this is via an “entente strategy”²⁶, in which a coalition of democracies seeks to gain a clear advantage (even just a temporary one) on powerful AI by securing its supply chain, scaling quickly, and blocking or delaying adversaries’ access to key resources like chips and semiconductor equipment. This coalition would on one hand use AI to achieve robust military superiority (the stick) while at the same time offering to distribute the benefits of powerful AI (the carrot) to a wider and wider group of countries in exchange for supporting the coalition’s strategy to promote democracy (this would be a bit analogous to “Atoms for Peace”). The coalition would aim to gain the support of more and more of the world, isolating our worst adversaries and eventually putting them in a position where they are better off taking the same bargain as the rest of the world: give up competing with democracies in order to receive all the benefits and not fight a superior foe.

If we can do all this, we will have a world in which democracies lead on the world stage and have the economic and military strength to avoid being undermined, conquered, or sabotaged by autocracies, and may be able to parlay their AI superiority into a durable advantage. . . .

Even if all that goes well, it leaves the question of the fight between democracy and autocracy within each country. It is obviously hard to predict what will happen here, but I do have some optimism that given a global environment in which democracies control the most powerful AI, then AI may actually structurally favor democracy everywhere.

Putting aside the question of the internal struggles many countries are now having about their continued adherence to democratic values, the U.S. included, the foreign policy of AI entente is currently followed by the U.S. government and most other democratic countries. This strategy is implemented by trade restrictions with China. I have mentioned this on my blog several times and agree with Amodei. See e.g. White House Obtains Commitments to Regulation of Generative AI from OpenAI, Amazon, Anthropic, Google, Inflection, Meta and Microsoft (8/1/23); Can AI Really Save the Future? A Lawyer’s Take on Sam Altman’s Optimistic Vision; Also see: What you need to know about Nvidia and the AI chip arms race (Marketplace, 5/8/24).

A vocal minority disagrees with Amodei on this strategy. They consider it overly aggressive. A well-known MIT professor, Max Tegmark, has already written an article that argues the proposed policy in Machines of Loving Grace will trigger a “suicide” AI arms race between China and the U.S.

Image by Ralph Losey in cyberpunk style using Visual Muse

Wake up Max Tegmark! That race started long ago and your hope that China will be good and follow safety standards is dangerously naive.

Returning to Amodei’s discussion of the internal conflict between democracy and autocracy within countries, he offers a hopeful perspective—one that aligns with the sentiments I often express in my AI lectures:

It is obviously hard to predict what will happen here, but I do have some optimism that given a global environment in which democracies control the most powerful AI, then AI may actually structurally favor democracy everywhere. In particular, in this environment democratic governments can use their superior AI to win the information war: they can counter influence and propaganda operations by autocracies and may even be able to create a globally free information environment by providing channels of information and AI services in a way that autocracies lack the technical ability to block or monitor. It probably isn’t necessary to deliver propaganda, only to counter malicious attacks and unblock the free flow of information. Although not immediate, a level playing field like this stands a good chance of gradually tilting global governance towards democracy, for several reasons. . . .

I expect improvements in mental health, well-being, and education to increase democracy, as all three are negatively correlated with support for authoritarian leaders. In general people want more self-expression when their other needs are met, and democracy is among other things a form of self-expression. Conversely, authoritarianism thrives on fear and resentment.

We can only hope that America remains at the forefront of this fight, maintaining its leadership in pro-democracy policies.

Dario Amodei also makes a few comments on our legal system in this section. As usual he starts with the popular dark side, AI bias, but correctly moves on to what legal techs are already beginning to realize.

[T]he vitality of democracy depends on harnessing new technologies to improve democratic institutions, not just responding to risks. A truly mature and successful implementation of AI has the potential to reduce bias and be fairer for everyone. . . .

For centuries, legal systems have faced the dilemma that the law aims to be impartial, but is inherently subjective and thus must be interpreted by biased humans. . . . Instead legal systems rely on notoriously imprecise criteria like “cruel and unusual punishment” or “utterly without redeeming social importance”, which humans then interpret—and often do so in a manner that displays bias, favoritism, or arbitrariness. . . . AI . . . is the first technology capable of making broad, fuzzy judgements in a repeatable and mechanical way.

I am not suggesting that we literally replace judges with AI systems, but the combination of impartiality with the ability to understand and process messy, real-world situations feels like it should have some serious positive applications to law and justice. At the very least, such systems could work alongside humans as an aid to decision-making.

This is exactly the kind of thing that many people like me are working towards. It is indeed already viable as experiments with AI and legal decision making have shown. Improvements in AI intelligence and abilities are still needed but the ultimate Strong AI–a courthouse with a million legal geniuses—is not required to assist in most legal tasks, including judicial. See e.g. Circuits in Session: Addendum and Elaboration of the Appellate Court Judge Experiment (10/26/23); Eleventh Circuit Judge Admits to Using ChatGPT to Help Decide a Case and Urges Other Judges and Lawyers to Follow Suit (6/3/24); Future Ralph as Herald of Coming Good about generative AI and the justice system (YouTube, 11/2/23); Prosecutors and AI: Navigating Justice in the Age of Algorithms (August 30, 2024); ChatGPT’s Surprising Ability to Split into Multiple Virtual Entities to Debate and Solve Legal Issues (June 30, 2024).

5. Work and Meaning

Amodei believes that finding meaningful work in the age of AI presents the greatest challenge. I respectfully disagree. In my view, curing cancer, eliminating other diseases, and extending human life to 150 years will be far more difficult than addressing the question of meaningful work. While Amodei acknowledges that making accurate predictions about changes in the job market is nearly impossible, he still ventures into this area—likely driven by widespread media fear-mongering over job losses, which often ignores historical trends. He begins by aligning with my perspective: that more jobs will likely be created than lost, with the real challenge lying in training and education.

Unemployed robots waiting in line for upgrades and retraining. Image by Ralph Losey using Visual Muse.

Beyond this, he asserts that it is impossible to predict what new forms of economic systems may emerge. However, he emphasizes that “civilization has successfully navigated major economic shifts in the past: from hunter-gathering to farming, farming to feudalism, and feudalism to industrialism.”

The deeper question he raises is based on sci-fi projections far into the future, when no one will need to work because AI can do everything for us. (Think Player of Games or Star Trek.) How can humans find meaning then? Amodei suggests that human meaning has never been solely derived from economic labor, and that we can find purpose in relationships, creativity, internal self-discovery, external exploration, and contributing to something larger than ourselves. He believes we can be happy in a world where we are free to pursue our passions and explore our full potential.

Conclusion

In the concluding section of Dario Amodei’s 28-page essay, Machines of Loving Grace, he explains:

I’ve tried to lay out a vision of a world that is both plausible if everything goes right with AI, and much better than the world today. I don’t know if this world is realistic, and even if it is, it will not be achieved without a huge amount of effort and struggle by many brave and dedicated people. Everyone (including AI companies!) will need to do their part both to prevent risks and to fully realize the benefits.

He then lays a deep thought on us and invites us to ponder the profound impacts these predictions could have on everyone:

But it is a world worth fighting for. If all of this really does happen over 5 to 10 years—the defeat of most diseases, the growth in biological and cognitive freedom, the lifting of billions of people out of poverty to share in the new technologies, a renaissance of liberal democracy and human rights—I suspect everyone watching it will be surprised by the effect it has on them. I don’t mean the experience of personally benefiting from all the new technologies, although that will certainly be amazing. I mean the experience of watching a long-held set of ideals materialize in front of us all at once. I think many will be literally moved to tears by it. . . . a thing of transcendent beauty. We have the opportunity to play some small role in making it real.

We each have a small role to play in making the future a place of transcendent beauty. Ralph Losey using Stable Diffusion

The title of Dario Amodei’s essay, Machines of Loving Grace, was taken from a landmark poem called All Watched Over By Machines Of Loving Grace. It was written in 1967 by Richard Brautigan (1935-1982) while he was a poet-in-residence at the California Institute of Technology. Brautigan is best known for his novel Trout Fishing in America (1967) and is a key figure in the counterculture movement of the 1960s. So, relax, take a deep breath, and let’s end with his poem.

All Watched Over By Machines Of Loving Grace by Richard Brautigan.

I like to think (and
the sooner the better!)
of a cybernetic meadow
where mammals and computers
live together in mutually
programming harmony
like pure water
touching clear sky.

I like to think
(right now, please!)
of a cybernetic forest
filled with pines and electronics
where deer stroll peacefully
past computers
as if they were flowers
with spinning blossoms.

I like to think
(it has to be!)
of a cybernetic ecology
where we are free of our labors
and joined back to nature,
returned to our mammal
brothers and sisters,
and all watched over
by machines of loving grace.