Promise
“AI will be the most transformative technology of the 21st century. It will affect every industry and aspect of our lives.” —Jensen Huang, CEO of Nvidia
Peril
“The development of full artificial intelligence could spell the end of the human race…. It would take off on its own, and re-design itself at an ever-increasing rate. Humans, who are limited by slow biological evolution, couldn’t compete, and would be superseded.” —Stephen Hawking, theoretical physicist and cosmologist
This is the third installment of the Shouting from the Soapbox series from my blog, Mixed Metaphors, Oh My!
“Pick a subject you care so deeply about that you’d speak on a soapbox about it.” — Kurt Vonnegut
A blog is many things, a public journal, a conversation with oneself, a showcase for writing and ideas, an exercise in vanity and a soapbox. Vonnegut’s quote speaks to me. I write about subjects I care deeply about, my relationship with myself, relationships with others, and my place in the larger community. I write from my lived experience, what I’ve learned from others, and what it means in the larger context of the world we live in.
During the past thirteen years since I’ve launched this blog, I’ve often stood up and shouted from my soapbox. I’ve often written that “The personal is political and the political is personal.”
Before I Begin
As a person of a certain age, 76-years-old to be precise, and a late adopter to technology, I don’t want to be left behind. As we age, many of us are already at risk for becoming isolated, invisible, simply because of age in our Western culture, and potential victims of scams and hacking, our personal information compromised.
The technology tools are changing every day and I’ve already experienced resistance or lack of trust in using some of them. I still prefer cash to debit cards, email to texts, Facebook and Messenger instead of Instagram, Tik Tok, Facetime, Discord, or Signal. Though I’m a visual person, I still love the printed word, not only images to convey my message. I still keep handwritten journals in addition to digital ones. I send greeting and thank you cards. Yes, I use snail mail and still have a checkbook.
I consume news and content by watching television, films, videos, and reading online content, including magazines like The Atlantic. I subscribe to blogs and Substack covering politics, culture, and health. I’m a lifelong reader and still purchase books and visit libraries.
For my blog, Mixed Metaphors, Oh My! I write about topics that capture my curiosity and reflect, or contrast, my personal POV, hoping I will find an audience that resonates with my work. Two years ago, I wrote my first essay on Artificial Intelligence, AI: “Be afraid, very afraid.” Earlier, this year I saw the documentary, The AI Doc: Or How I Became an Apocaloptimist. It inspired me to keep reading on the topic, become more informed, and weigh the promise vs. peril to the best of my ability. I’ve paraphrased content, including comments and concerns from the documentary.
An excerpt from my blog post, AI: “Be afraid, very afraid.”
“What, me worry?” Is the catch phrase of Mad Magazine’s, imaginary character, Alfred E. Neuman. When reading the satire and comic genius of the magazine growing up, the answer for most boomers, was no, I’m not worried, until it quickly changed after the proliferation of nuclear weapons, fallout shelters, duck and cover civil defense drills, the Kennedy assassination and the list goes on.
Television from that era, The Twilight Zone, and Outer Limits, played on that paranoia and dis-ease. Some of the most popular films of the 1960’s and continuing today, ask the questions we are afraid to answer.
Humans are now at a crossroads — we now need to worry whether the tools like AI, which we’ve created to assist us — now in the end — may be our final undoing.
Writer’s Statement on AI Use:
AI does not draft content for this blog. The words, sentence construction, grammar, spelling — and most of all the message and point of view — are mine. Full disclosure: I hold no writing degrees or credentials, though I studied journalism and communication arts at the University of Wisconsin. I’ve been writing for over 60 years, both personally and professionally. I became a writer by writing.
For my blog, Mixed Metaphors, Oh My! I write and publish poetry, reminiscences, spoken word monologues, and essays on culture and politics. I continue to work on my memoir, Perfectly Flawed and journal and post on social media.
I use AI for reference and research and will note it. NOTE: For this essay I’ve used AI tools to research articles, select images, view videos and other content, spell check names, fact check sources, and additional use as noted.
My writing style reflects the way I tell stories, run-on sentences, favorite misspelled, and sometimes wrong words and phrases, euphemisms, and ‘the list goes on’, the latter a perfect example! I often meander in nonlinear directions. Yes, I sometimes get lost when writing like storytelling in person.
In essence, my writing style is my voice.
About This Essay
My intention is to first, educate myself, second, share what I’ve learned, and lastly, attempt to begin to analyze the
promise vs. peril of AI. It’s naïve of me to believe I have the answers, I have more questions than answers. What I’ve learned so far is that AI is smarter, faster, more powerful and dangerous than I am. My job is to learn how to use it responsibly while protecting myself from harm and/or misinformation.
Following is a glossary of terms, top AI tech companies and leaders, and a poll on how voters weigh in on the risks vs, benefits of AI. At the end of this essay is related reading from Mixed Metaphors, Oh My! and additional content from other sources as noted.
Full disclosure: The content that follows is generated by AI and reprinted for educational purposes. It’s an example of how I currently use AI as a tool. I encourage readers to fact check and explore on your own.
At the end of this essay, I’ll share my takeaways on the promise or peril of Artificial Intelligence.
Artificial Intelligence Glossary of Terms
Note: The following glossary is reprinted from a source article from MIT Sloan Teaching & Learning Technologies.
At a Glance
Artificial Intelligence (AI) stands at the forefront of technological advancement, shaping our daily interactions and revolutionizing industries. We’ve created this glossary to help you build a foundational understanding of generative AI tools.
Whether you’re a newcomer or an AI veteran, learning the basic vocabulary behind these technologies can help you gain a better understanding of AI tools’ opportunities and subtleties. Developing literacy in AI concepts will also enable our community stay at the forefront of technological advancement. We can lead nuanced conversations about balancing innovation with ethical considerations and help steer AI toward positive impact.
This glossary is inspired by the New York Times Artificial Intelligence Glossary and clarifies essential terms related to the generative AI landscape here at MIT Sloan.
Anthropomorphism
We use the term anthropomorphism to describe the habit of assigning human-like qualities to AI. While AI systems can imitate human emotions or speech, they don’t possess feelings or consciousness. We might interact with various AI models as if they were colleagues or thought partners, but in reality, they serve as tools for learning and resource development.
Agents
Agents are autonomous or semi-autonomous AI entities that can perform tasks, make decisions, and call tools or APIs based on goals. In academic and enterprise settings, agents are often used to automate workflows like document summarization, task routing, or multi-step reasoning.
Bias
Bias in AI models refers to output errors caused by skewed training data. Such bias can cause models to produce inaccurate, offensive, or misleading predictions. Biased AI models arise when algorithms prioritize irrelevant or misleading data traits over meaningful patterns (Smith, 2019).
Chain-of-thought Prompting
Chain-of-thought prompting is when you write prompts to encourage the AI model to reason step-by-step before arriving at an answer. This technique can improve the accuracy and applicability of AI output (Wei et al., 2022). It’s especially useful for math problems, logic, or any multi-step decision-making.
Context window
The context window is the maximum number of tokens (words or parts of words) that an AI model can process and consider simultaneously when generating a response. It is essentially the “memory” capacity of the model during an interaction or task. Models with larger context windows can handle larger attachments/prompts/inputs and sustain “memory” of a conversation for longer (Fogarty, 2023).
Emergent Behavior
We call the unexpected skills showcased by vast language models emergent behaviors (Pasick, 2023). These talents span coding, musical composition, poetry crafting, and even the creation of fictional narratives.
Fine-tuning
Further training a pre-existing AI model on a specialized dataset to improve its performance on specific tasks, audiences, or domains.
Generative AI
Generative AI is an advanced technological approach that enables the creation of content including text, images, and videos. By analyzing and discerning patterns within extensive training datasets, generative AI can autonomously construct material that shares comparable characteristics to its training input. This capability stems from the AI’s understanding of data patterns and its ability to replicate or innovate based on these patterns.
Whether it’s generating art, writing prose, or crafting other digital content, generative AI leverages its learned knowledge to produce results that often mirror human-like creativity. While generative AI systems may seem human in nature, they do not possess human consciousness or emotions themselves.
Hallucination
We call the occurrences where large language models generate factually inaccurate or illogical answers due to data and architecture constraints hallucinations.
Large Language Model (LLM)
Neural networks known as large language models work by forecasting word sequences. Large language models’ capabilities are rapidly advancing and continue to evolve with increased use. They can now hold dialogues, write prose, and scrutinize enormous text quantities from the internet.
Meta Prompt / System Prompt
A meta prompt, also called a system prompt, is a set of instructions provided to the AI model behind the scenes before user interaction begins. These instructions may be hidden from the user. Meta prompts set behavior, tone, or boundaries for how the AI should respond (e.g., “You are a helpful teaching assistant”).
Multimodal Model
A multimodal model is an AI model capable of processing and generating multiple types of input/output such as text, images, audio, and video. Multimodal tools, such as those offered by OpenAI, Anthropic (Claude), and Google (Gemini), can process both text and images within the same conversation. For example, you could upload a diagram and ask the model to explain it or generate code from it. Some models, like Google Gemini, also support audio and video input.
Natural Language Processing (NLP)
Natural Language Processing (NLP) is a subfield of artificial intelligence and computational linguistics that focuses on enabling machines to understand, interpret, and generate human language to be understood by humans.
Neural Networks
Neural Networks, modeled after the human brain, are a mathematical system that actively learns skills by identifying and analyzing statistical patterns in data. This system features multiple layers of artificial neurons, which are computational models inspired by the neurons in our brain
These artificial neurons process information and transmit signals to other connected neurons. While the first layer processes the input data, the final layer delivers the results (Hardesty, 2017). Intriguingly, even the experts who meticulously design these neural networks often find themselves puzzled by the intricate processes occurring between the layers.
Parameters
In the realm of AI systems, developers establish numerical values referred to as parameters. For context, OpenAI’s GPT-4 is believed to incorporate hundreds of billions of parameters that drive its ability to predict words and create dialogue. Consider these two parameters, which play a pivotal role in shaping both the construction and behavior of a large language model:
- The construction parameter refers to the underlying structure and architecture of the model. This includes how layers of artificial neurons are organized, interconnected, and weighted. It’s akin to the framework or skeleton that gives shape to the model.
- The behavior parameter refers to how the model operates, reacts, and evolves in response to input data. It defines the model’s responsiveness, adaptability, and its specific output patterns. The behavior can vary based on factors such as the type of input data and external connectivity, like internet access.
Prompt Engineering
Prompt engineering is the practice of designing effective prompts to guide an AI model’s output. This involves setting roles, specifying format, adding constraints, or giving examples to improve the quality, tone, or relevance of the response.
RAG (Retrieval-Augmented Generation)
RAG (Retrieval-Augmented Generation) is a method that combines a language model with external sources added by the user, such as documents, PDFs, or other materials. While language models can generate clear and human-like responses, they don’t automatically have access to this added content. RAG retrieves relevant information from those sources, allowing the model to give more accurate and grounded answers.
Reasoning Models
A class of LLMs that think through complex problems step by step before generating a response, improving accuracy on logic, math, and multi-step analysis tasks.
Reinforcement Learning
Reinforcement Learning is a method in AI training where models learn optimal decision-making strategies through cycles of actions and feedback, with human interaction playing a pivotal role in refining the learning process. Models learn by making decisions, observing the outcomes of those decisions, and adjusting their strategies accordingly.
Temperature (AI Temperature)
An AI tool’s temperature setting controls how deterministic or creative that AI model’s output is. Lower values (e.g., 0.2) lead to more focused and consistent answers, while higher values (e.g., 0.8) produce more varied or imaginative responses.
Token
A token is the smallest unit of text that an AI model processes and understands; this is typically 4 characters in English, or about ¾ of a word. Tokens may include whole words, parts of words, individual characters, punctuation marks, and special characters (LinkedIn Learning, n.d.).
Transformer Model
Transformer models can process entire sentences simultaneously rather than in sequence, aiding in grasping context and the language’s long-term associations. This means these models can detect and interpret relationships between words and phrases in a sentence, even if they are positioned far apart from each other.
The Top AI Leaders & Tech Companies
Note: This list and content were generated by AI.
Sam Altman
As the CEO of OpenAI, Altman has been one of the primary drivers of the generative AI boom. He champions the long-term vision of developing Artificial General Intelligence (AGI) and integrating human-like reasoning models into public and enterprise workflows.
Elon Musk
Musk, the CEO of xAI, focuses heavily on creating safe, truth-seeking, and curious AI systems. He approaches the development of artificial intelligence as a pivotal step toward the technological singularity.
Mark Zuckerberg
As the CEO of Meta, Zuckerberg is a major proponent of open-source artificial intelligence. Through initiatives like the Meta Llama family of models, his goal is to democratize intelligence, drive adoption across social platforms, and integrate AI into spatial computing and AR/VR ecosystems.
Dario Amodei
Amodei is the CEO and co-founder of Anthropic, which develops the Claude series of language models. His leadership places a heavy emphasis on AI safety, alignment, and mitigating potential societal and security risks associated with advanced machine learning.
Sir Demis Hassabis
Hassabis is the CEO and founder of Google DeepMind. He leads the charge in merging artificial intelligence with scientific discovery, responsible for major breakthroughs in fields like biology (AlphaFold) while overseeing Google’s massive global AI platforms.
How Do Voters Weigh in on the Risks vs, Benefits of AI
Reprinted from an NBC Poll in March, 2026:
Voters are worried about AI and don’t trust either political party to handle the rapidly evolving
technology, according to a new national NBC News survey.
A majority of registered voters, 57%, said they believe the risks of AI outweigh its benefits, compared with 34% who said the opposite. What’s more, a plurality of voters view AI negatively and don’t believe either Democrats or Republicans are doing a good job handling policy related to the rapidly advancing technology.
Just 26% of voters say they have positive feelings about AI, compared with 46% who hold negative views. In fact, the only topics with a lower net positive rating than AI in the NBC News survey were the Democratic Party and Iran.
How voters feel about political figures and topics
AI had the third-lowest net positive rating of the group.
Notes: The poll was conducted Feb. 27-March 3 and surveyed 1,000 registered voters nationally. The margin of error is plus or minus 3.1 percentage points. Source: NBC News poll. Graphic: NBC News
What Have I Learned?
The Promise
First, this is what I’ve learned so far about AI predictions, the potential and promise of how it could enhance our lives:
What is Artificial Intelligence?
- Tech experts and ethicists have described it as a ‘magical computer box,’ making predictions. Its intelligence is in recognizing patterns, and making a prediction on the next word in a sentence.
- AI analyzes data and teaches itself, rather than being programmed, uncovering the patterns of the universe.
- Note: The following is an AI generated summary:
Artificial General Intelligence (AGI) is a hypothetical type of AI that can understand, learn, and apply knowledge across a wide variety of intellectual tasks at a human level or higher. Unlike today’s narrow AI, which is specialized for specific jobs, AGI would possess broad reasoning, adaptability, and “common sense.”
Tech leaders describe the promise of AI:
- It is the most consequential movement since the industrial age.
- AI will do all of the research for us, faster, smarter, and become super human.
- Greatest impact on school.
- It’s designed to do everything better than humans.
- Potential to automate all human labor.
- Ultimately it will be smarter than all of humanity.
- Technology will turn scarcity into abundance. Create an ‘age of abundance.’
- Can solve pandemics, help with climate change.
- Provide access to healthcare.
- Extend ‘healthspan’ not just lifespan.
- One tech leader described the future potential and promise of AI as, “The only day better than today is tomorrow.
The Peril
- As AI replaces some jobs, there’s currently no plan for a life without a job or income. The advancements outpace plans for the future.
- AI requires more resources than we’ve ever spent before.
- Data centers use millions of gallons of water each day, and massive amounts of electrical power, plus the land they sit on.
- Many experts believe AI has been deployed without due diligence. Some critics claim, it’s in the tech companies’ best interest to mislead the public.
- Generative AI could flood the world with disinformation.
- Runaway AI development by a small number of tech companies can create a concentration of wealth and power, a ‘profit maximization incentivization.’
- Some describe AI as an authoritarian’s dream. AI tools can scale up to a level of authoritarianism.
- The fear many express is that whoever wins the AI race, will govern the world.
- An additional fear, in an effort to win the race, safe deployment is jeopardized.
- Worries about geopolitical competition. The U.S and China are in an AI arms race.
- Five CEO’s control AI in the U.S with affiliated partnerships and corporations.
- We need to create oversight (third party evaluation), accountability (transparency), and regulation (legally liable).
Civic Engagement
Note: This list and content were generated by AI.
Societal and Ethical Implications
The pursuit of AGI comes with significant societal debates, as its realization could fundamentally disrupt every knowledge-based industry:
- The Alignment Problem: Ensuring that AGI systems remain aligned with human values and do not develop goals contrary to human well-being.
- Economic Impact: The potential for widespread labor displacement and the need to redefine human work, prompting discussions about policies like Universal Basic Income (UBI).
- Existential Risk: While some leaders advocate that preventing human extinction via AI should be a global priority, others believe AGI is too theoretical in the near term for such existential concerns.
How Do I Weigh-In on the Promise vs. Peril of AI?
I return to the beginning of this essay. Like most things in life that are important, I have more questions than answers, more learning ahead. It’s my responsibility as a citizen of the world to be responsible, honest, ethical and to the best of my ability as I use these tools, to be transparent when AI assists me while I strive as a write to protect my voice.
In the end the what I believe from the core of my being is that, “The personal is political and the political is personal.” I will do my best to be open to the promise of AI and vigilant regarding the peril.
Related Reading from Mixed Metaphors, Oh My!
Shouting from the Soapbox: New Series
Shouting from the Soapbox: Russian Roulette
Additional Content on the Topic
From The Atlantic:
How to Think About AI Before It’s Too Late
The People Who Will Thrive in the AI Age
Inside the Dirty, Dystopian World of AI Data Centers
Would Claude Refuse an Illegal Military Order?
The AI Super PACs Trying to Influence the Midterms
From other content sources:
The AI Doc: Or How I Became in Apocaloptimist
AI could breach government and business defenses in months, US and its intelligence partners warn
US government allows Anthropic limited release of AI model that sparked cybersecurity
AI’s Empire: The Limits of Knowledge, and Predicting the Job Future
Poll: Majority of voters say risks of AI outweigh the benefits