Large Language Models (LLMs) are naturally influenced by their training data. Any biases present in the training data, whether intentional or unintentional, will naturally creep into the responses that the LLMs provide.
If I may take an extreme example (and prove Godwin’s Law in the process)…had Hitler developed an LLM in the late 1930s, you can imagine how it would answer selected questions about nationalities, races, or ethnic groups.
Of course that has nothing to do with the present day.
Red LLM, blue LLM?
But what IS newsworthy is that despite the presence of many technology leaders at President Donald Trump’s inauguration, I am unable to find any reference to a “red LLM.” Or, for that matter, a “blue LLM.”
Take the red LLM or the blue LLM.
Perhaps the terminology isn’t in vogue, but when you look at algorithmic bias in general, has anyone examined political bias?
Grok and bias
One potential field for study is Grok. Of all the godfathers of AI, Elon Musk is known both for his political views and his personal control of the companies he runs.
“Specifically, Grok falsely claimed that Kamala Harris, the Democratic presidential nominee, had missed ballot deadlines in nine states—an assertion that was entirely untrue.”
Yes, it sounds bad—until you realize that as recently as January 2025 some Google AI tools (but not others) were claiming that you had to tip Disney World cast members if you want to exit rides. Does Alphabet have a grudge against Disney? No, the tools were treating a popular satirical article as fact.
What data does Grok use?
“Grok is trained on tweets—a medium not known for its accuracy—and its content is generated in real-time.”
Regardless of how you feel about bias within X—and just because you feel about something doesn’t necessarily mean it’s true—the use of such a limited data set raises concerns.
But data isn’t the only issue. One common accusation about Grok is that it lacks the guardrails that other AI services have.
No guardrails.
A little secret: there are several reasons why Bredemarket includes wildebeest pictures, but one of them is that my version of Google Gemini does not presently generate images of people because of past image generation controversies.
“grok 2.0 image generation is better than llama’s and has no dumb guardrails”
Whether a particular guardrail is good or bad depends upon your personal, um, bias.
After all, guardrails are created by someone, and guardrails that prevent portrayal of a Black President, a man with a U.S. (or Confederate) flag wearing a red cap, or an independent Ukraine or Israel would be loved by some, unloved by others.
In essence, the complaints about Grok aren’t that they’re biased, but that they’re unfettered. People would be happy if Musk functioned as a fetterman (no, not him) and exerted more control over the content from Grok.
But Musk guardrailing Grok output is, of course, a double-edged sword. For example, what if Grok prohibited portrayal of the current U.S. President in an unfavorable light? (Or, if Musk breaks with Trump in the future, in a favorable light?)
Not that it’s authoritative, but I asked Google Gemini what may happen to TikTok users on Sunday, January 19. The text below was written by Gemini.
On Sunday, January 19th, 2025, if the current legal situation remains unchanged:
New users will not be able to download TikTok.
Existing users will likely still be able to use the app on their devices. However, the app will no longer receive updates or bug fixes, potentially leading to performance issues and security vulnerabilities over time.
Important Note: This is based on the current understanding of the legal proceedings. The situation could change if:
The Supreme Court intervenes: They could potentially block or delay the ban.
TikTok is sold: If TikTok is sold to an approved buyer, the ban may be lifted.
I recommend staying updated on the latest news and legal developments regarding this issue.
When I tried to do so recently, I got an error from Google, saying in part that YouTube on TV’s request does not comply with Google’s “Use Secure Browsers” policy.
It’s worth noting:
The YouTube app on my phone is from Google.
The YouTube on TV app on my TV is from Google.
So if I read the message as stated, Google’s apps don’t meet Google’s standards.
(And for those who are curious, no my wi-fi network is not tied to my consulting e-mail account.)
I was in an impatient mood and didn’t feel like diagnosing the issue, so I didn’t.
“Plug and play” has a way to go in the wi-fi world.
Upload PDFs, websites, YouTube videos, audio files, Google Docs, or Google Slides, and NotebookLM will summarize them and make interesting connections between topics, all powered by Gemini 1.5’s multimodal understanding capabilities.
With all of your sources in place, NotebookLM gets to work and becomes a personalized AI expert in the information that matters most to you….
Our new Audio Overview feature can turn your sources into engaging “Deep Dive” discussions with one click.
I uploaded the most recent version of my resume to NotebookLM.
Technically, this is not my resume; this is a PDF version of a portion of my LinkedIn profile. But my resume has similar information.
NotebookLM used the resume as source material to create a 20+ minute podcast called “Career Detective.” In the podcast, a male and a female pair of bots took turns discussing the insights they gleaned from the resume of John E. “Breedehoft.” (I use a short e, not a long e, but people can call me anything if I get business from it.)
Surprisingly, they didn’t really hallucinate. Or at least I don’t think they did. When the bots said I was deeply qualified, as far as I’m concerned they were speaking the truth.
They even filled in some gaps. For example, I used the acronyms for KYC, KYB, and AML on my resume to save space, so one of the bots explained to the other what those acronyms meant, and why they were important.
Probably the most amusing part of the podcast was when they noted that I had worked at two very large companies. (Just so you know, my resume only goes back to 2015, so Motorola isn’t even discussed.) While Incode and IDEMIA are both multinationals, I wouldn’t characterize Incode as massive.
Anyway, judge for yourself
So here’s the audio episode of “Career Detective” that focuses on…me.
By the way, I learned about NotebookLM via the Never Search Alone Slack workspace, but still need to explore NotebookLM’s other features.
I have a telehealth appointment next week with a medical professional whom I have previously met. And I assume she will participate in the telehealth appointment.
In the future, of course, she may not.
Way back in April 2013, I wrote a tymshft piece entitled “You will still take a cab to the doctor’s office. For a while.” It speculated about a future 2023 medical appointment in which the patient took a driverless cab to a medical facility. In the office, the patient was examined by remote staff…or so she thought.
“Well, I’m glad you’ve gotten used to the procedure,” replied the friendly voice. “I hope you like me!”
“I do,” said Edith. “You’ve been very helpful. But I’ve always wondered exactly WHERE you were. If you were in Los Angeles, or in Mississippi, or perhaps in India or China, or perhaps even in one of the low-cost places such as Chad. If you don’t mind my asking, exactly where ARE you?”
“I don’t mind answering the question,” replied the friendly voice, “and I hope you don’t take my response the wrong way, but I’m not really a person as you understand the term. I’m actually an application within the software package that runs the medical center. But my programmers want me to tell you that they’re really happy to serve you, and that Stanford sucks.” The voice paused for a moment. “I’m sorry, Edith. You have to forgive the programmers – they’re Berkeley grads.”
“Oh,” said Edith after a moment. “This is something new. I’m used to it in banking, but I didn’t realize that a computer program could run an entire medical center. Well…who picks up the trash?”
“That’s an extra question! Just kidding,” replied the friendly voice. “Much of the trash pickup is automated, but we do have a person to supervise the operation. Ron Hussein. You actually know him – he was your cab driver in 2018 when you came here.”
Re-reading this 2013 piece, I was amused at three things I got wrong.
First, Google, Facebook, and Apple did NOT merge to form Gaceapple, “the important merger that saved the tech industry in the United States from extinction.” American tech firms are still powerful…for now.
Second, my assumption of cab companies adopting driverless cars assumed the continued existence of cab companies. Ride share services have reduced the presence of traditional companies dramatically.
Third, my assumption that medical firms would sink untold sums of money into centralized automated medical examination rooms could be questioned…especially for routine appointments like Edith’s. Why not just let Edith’s smartphone—perhaps with a single attachment—gather the data?
Of course, there are medical ethics questions that underlie this entire discussion of remote telehealth and the use of non-person entities (NPEs). And we are struggling with those right now.
The Google Business Listing service sellers are almost as bad as the 17x professional resume writers.
A bit of history: when I started Bredemarket I also set up a Google Business Listing, which also included a free ugly Google website to promote my business. I poured my energies into bredemarket.com instead.
Bredemarket’s geographic market as of April 2021, according to my Google business listing page. The old link now redirects to a general Google search for Bredemarket.
I’d occasionally add content to my Google Business Listing, but sometimes Google would flag the content for some unknown reason, and eventually Google flagged the entire Google Business Listing for some unknown reason. (Google: “You violated our TOS, but we won’t tell you why. You figure out what you did.”) By this point I was happy to be rid of the thing.
But I’m still getting calls at least once a week from companies that want me to get my Google Business Listing properly verified. When I tell these companies that losing my Google Business Listing was the best thing that ever happened to me, they usually hang up.
Usually.
My “Google” review (which wasn’t about me, and wasn’t on Google)
Today the caller was persistent, mentioning at one point that Bredemarket had five star reviews saying “I was blown away with this service”…from a resume writing service. I asked for the date of the review, and the caller couldn’t tell me. And you know how I feel about some resume writing services anyway.
After this caller finally gave up on selling their services, I searched Google for a five-star review of Bredemarket.
I ended up at Bredemarket’s Yelp page…which has no reviews. But if you scroll down, you see sponsored content about OTHER websites…including KStar Professional Resume Writing Services, for which a reviewer wrote “I was blown away with this service.”
So instead of a five star Google review of Bredemarket, there is a five star review of a completely separate resume writing service that appears in the sponsored ads for Bredemarket’s Yelp page.
My current two categories for go-to-market collateral, “external” and “internal,” are not sufficient. I want to fix that, but I’m still thinking through things, such as what to call the third category. As of this moment, my best option is “intrinsic,” based upon my conversations with my good buddy Gemini.
My current two categories for go-to-market collateral
Based upon go-to-market efforts that stretch back a decade (even before I formally became a Product Marketing Manager), I have traditionally divided go-to-market output into two categories.
22 types of content Bredemarket can create.
External content for your hungry people (target audience), such as articles, blog posts, social media, case studies, white papers, and proposals. This content goes to CEOs, marketers, engineers, IT staff, and many others at the companies that buy from you.
Internal content for the people in the company who talk to your hungry people, such as battlecards, sales playbooks, and scripts for trade show demonstrations. This content goes to salespeople, marketers, product managers, customer success folks, and many other employees and contractors.
Some content can either be external or internal, depending upon the audience.
Simple and straightforward, right?
My emerging third category
But during a current go-to-market effort for a client, I realized that these two categories are insufficient, since both the client and I are creating content that will never been seen by either the external clients or the internal clients.
I’ll give you two heavily redacted and anonymized examples for a #fakefakefake machine.
AI-generated by Google Gemini.
Bob wrote a Word document that listed the pain points that MegaCorp’s client encounter when trying to put green round widgets on their end customers’ beanie caps.
Judy created a picture with stick figures that showed how clients can use the MegaCorp WidgetPlacer 2000 for their beanie cap production needs.
Bob and Judy (names from this song) are NEVER going to show that document or that picture to the customers, or even to the company’s own account managers.
So why create it?
Because they will be the initial guides as the go-to-market team creates all the other external and internal collateral for the effort.
These foundation documents are a critical part of the go-to-market effort. Without them, the subsequent go-to-market collateral will be a contradictory mishmash of stuff, and you’ll need to undertake a lot of effort to align all the content after the fact.
And the third category is…
…well, I don’t know what to call this third category yet.
“Internal internal content” seems a bit silly.
“Foundational content” seems a bit grandiose.
“Bob and Judy” content seems like an independent TV show.
If we call the first category “external” marketing content for customers, and we call the second category “internal” marketing content for employees, then what do we call the third category which is only for the people who create marketing content for customers and employees?
Gemini’s first batch of answers:
Internal Marketing Content for Marketing Teams
Marketing Team Communication
Marketing Team Resources
Internal Marketing Tools
Brevity is NOT a strength of generative AI. Even when I specifically instructed Gemini to provide single word responses, I got hyphenated words:
Intra-marketing (combining “intra” for within and “marketing”)
Meta-marketing (implying a higher-level perspective on marketing)
Team-marketing (emphasizing its focus on the marketing team)
The fatal flaw is that not all the people on a go-to-market team are marketers. (This is my fault because of the prompt I wrote.)
And if I lop off the last part of each compound phrase, two of my options are “team” (too generic) and “meta” (NO).
And “intra” is too close to “internal internal.” But perhaps it’s a start. So I asked Gemini for words similar to “intra-” and got this response:
The prefix “intra-” means “within,” “inside,” or “inner.” Here are 12 words that convey a similar meaning:
Inward: Turning or moving toward the inside.
Internal: Existing or occurring within something.
Inner: Situated or located within.
Intrinsic: Belonging to or inherent in the nature of something.
Interior: The inside part of something.
Intramural: Occurring within a school or institution.
Intranet: A private computer network within an organization.
Intravenous: Administered into a vein.
Intracranial: Situated or occurring within the skull.
Intracellular: Existing or occurring within a cell.
Intralingual: Relating to or occurring within a language.
Intrajugular: Situated or occurring within the jugular vein.
I kind of like “intrinsic,” but I’ll sleep on it.
In the meantime, what would YOU call the third category?
Bredemarket 