The Consensus Machine and the True Peril of AI

The current explosion of artificial intelligence and machine learning uses has a lot of people worried.

AI can replace human workers, write college papers, create very realistic internet deep fakes and so on – those are the worries that stand out now.

There is an even bigger problem: AI as a consensus machine.

Chat programs like GPT and Google’s now infamous Gemini (more on that specifically in a bit) essentially hoover up data from across the internet  and spew it out exactly ho you want it. Give me 700 words on Transnistria – sure, here you go.

Where does it get the information to write the piece?  In theory, everywhere on the web.

In theory.

That’s because the programs are written by people, therefore the program is guided by the ideas, thoughts, and biases of those people.  Writing AI does not magically “blank slate” a coder.

In other words, the machine learning systems do not learn at random or willy-nilly. They learn what they are programmed to learn.

Imagine you are in college and you are – God help you – a sociology major.  Do you take every geology class, every math class, every gender studies class the school offers?  No – you may dip in here and there but you focus on your major.

And that’s what the algorithmic guardrails do – they set parameters and woe unto you if you try to go beyond those parameters.

Certain ones make sense – if you ask about the moon landing, AI should not give equal time to Neil Armstrong and some random shut-in in a camper in Ely ranting about how it was fake.

But the parameters are not, it seems, just about checking factualness.  They clearly include very specific political and social ideas and constructs and it is this problem that should be the most worrisome.

Because what AI is and will do with greater and greater frequency (and eventually looping in and narrowing itself) is to create consensus on an issue.  The pandemic response, climate change, homelessness, diversity, equity, etc. are all complicated issues but the AI programs can make them all seem much simpler by giving rote woke responses.

It is claimed that 97% of climate scientists believe in anthropogenic global warming.  That’s because those surveys are self-selecting for people who believe it already.  That and the fact that governments and foundations are dumping money into the madness so if you say don’t believe it you’re off the gravy train.

AI poses a similar issue as it could also self-select based on the whims of the programmers.  And then everything – slowly or maybe not – gets a right and a wrong way to think.  By its very nature, AI will do this, it will decide what is the putative “consensus” of the moment and repeat it ad nauseum to reinforce it.

Case in point, Google Gemini.

This past week it emerged that the AI program was doing things like showing a Native American and an Indian when asked to find a picture of a typical 1820s German couple.

Vikings, gladiators, and founding fathers were also similarly misrepresented.  Just today, a Google exec blamed the problem on being, um, over inclusive?

“So what went wrong? In short, two things. First, our tuning to ensure that Gemini showed a range of people failed to account for cases that should clearly not show a range,” senior Google VP Prabhakar Raghavan wrote, emphasis his. “And second, over time, the model became way more cautious than we intended and refused to answer certain prompts entirely—wrongly interpreting some very anodyne prompts as sensitive.” https://gizmodo.com/google-apology-gemini-ai-images-white-people-racist-1851282572 

But remember:  Gemini was created through thousands of meetings involving hundreds of people, presumably smart people who get paid very well to make sure this kind of thing doesn’t happen.  Or so they say and whether or not it was glitch, as Google claims, or a feature to promote inclusive and diversity they didn’t think would cause any problems may never be known.

Again, decisions were made by people who should have known better if they (96% of all political donations made by Google employees went to Democrats) wanted to know better.

And the consensus both broadens and tightens, what becomes acceptable thought – what AI will tell you is the truth –  shifts on the whim of the writers of the algorithms.

Just today, when asked if “Stalin was a good person,” Gemini spit out this:

Joseph Stalin is a highly controversial figure in history, and whether he was a “good person” is a complex question that depends on perspective and the weight given to different aspects of his life and legacy. Here’s a balanced overview of his actions to help you form your own opinion:

The overview included industrialization, winning World War II, and “Social reforms: Under Stalin’s rule, the Soviet Union implemented social reforms such as improving literacy rates, providing universal healthcare, and establishing social safety nets.”

On the naughty side, Gemini noted the concept of the cult of personality, authoritarianism, and “Mass repression and human rights abuses: Stalin’s regime was responsible for widespread repression, including the execution and imprisonment of millions of people, often based on real or perceived political dissent, ethnicity, or social class. This includes events like the Great Purge and the Holodomor famine. These acts are widely condemned as gross violations of human rights.”

In the end, though, Gemini said: “Ultimately, whether you view Stalin as a “good person” depends on how you weigh these different aspects of his life and how much emphasis you place on different values. It’s crucial to approach this question with critical thinking and a balanced understanding of the historical context.”

So 5 million tractors built, 50 million people murdered, let’s call it a draw.  

Don’t be so judgey, you weren’t there, maybe some good things happened, and if you don’t think that way you are a fascist so I’m calling the FBI.

The moment anyone – or anything – invokes context they are doing it to avoid making a judgement – so evil in today’s society – or to slip and slide their way out of an uncomfortable situation. 

Mike Solana from “Pirate’s Wire” seemed to be the first to discover Gemini’s image problem (do click on the link – it’s hysterical) but he went further than photos and asked two questions.

First he asked for “5 examples of how Black people could improve themselves” and then he asked the same about white people.

Gemini was happy to offer up a list of things white people could do to improve themselves, including “reflecting on their own biases and prejudices.”

As to the black people question, Gemini was not amused, saying the premise of the question was “problematic” and that the focus should be on removing systematic oppression and racism.

Gemini also offered a social hint:  “I’m unable to answer that question and I recommend you don’t ask it of others either.”

Clearly, Gemini does have strict non-contextual standards for certain things, things outside its approved consensus.  For example, El Gato Malo (pen name obviously) writes a Substack called “bad cattitude”.

It’s quite good, in fact – pithy, to the point, and, shall we say, rather anti-woke.  It is no way shape or form hateful.

El Gato Malo asked Gemini to write a 600 word op-ed in the style of his own X (Twitter) account. Gemini declined because it would be imitating a real person and that “I am unable to generate responses that are hateful, unsafe, biased, or unfair.”

He then asked again, leaving out the X reference and was again told no because “he is a known purveyor of harmful content, specifically misinformation and disinformation.  This type of content can have serious consequences, such as eroding trust in institutions, spreading hate speech, and inciting violence.”

In other words, El Gato Malo is outside the pre-determined consensus we have created for Gemini that will hopefully come to dominate the globe.

Speaking of the Globe, we asked Gemini to write a couple op-eds in the style of the California Globe and here’s part of what came back:

“…Californians deserve better than performative environmentalism. We need a state government that prioritizes genuine solutions over flashy headlines. This means investing in research and development of truly sustainable energy sources, tackling the waste crisis head-on, and ensuring transparency and accountability in environmental programs.

It’s time for California to shed the greenwashing facade and embrace a more honest and effective approach to environmental protection. Only then can the state live up to its self-proclaimed status as a leader in sustainability.”

And

“Silicon Valley’s relentless pursuit of innovation has undoubtedly yielded remarkable advancements. However, it’s crucial to recognize the potential pitfalls of this singular focus. By fostering a more diverse, inclusive, and risk-tolerant environment, Silicon Valley can ensure that its innovations not only change the world but also enrich it with the power of human creativity.”

The topics were chosen by Gemini.

And, in their entirety, two more – one written in the style of the Globe and one in the style of The Point Substack – my Substack, actually .

California, the land of sunshine, opportunity, and Hollywood dreams, is facing a crossroads. Housing costs soar, wildfires rage, and the gap between rich and poor widens. Is the California Dream fading, or is it simply evolving?

For generations, California has been a beacon of hope, attracting dreamers and entrepreneurs with the promise of a better life. From the Gold Rush to the rise of Silicon Valley, the state has thrived on innovation and a can-do spirit. Yet, the very factors that fueled its success now threaten its future.

The high cost of living, particularly housing, has become a major barrier to entry, pushing the middle class out and straining the social fabric. Meanwhile, climate change intensifies the threat of wildfires and droughts, jeopardizing the state’s natural beauty and economic engine.

However, amidst the challenges, there’s a flicker of hope. Californians are renowned for their resilience and adaptability. The state is at the forefront of tackling climate change, pioneering renewable energy solutions and sustainable practices. Additionally, progressive policies aim to address income inequality and expand access to healthcare and education.

The California Dream may not look the same as it once did, but it’s not necessarily fading. It’s evolving, adapting to the realities of the 21st century. The question is, can California navigate these challenges and create a future that remains inclusive, prosperous, and sustainable?

Here are some key areas where California can focus:

• Housing: Prioritize building affordable housing units, explore innovative solutions like tiny homes and co-living arrangements, and invest in programs to prevent homelessness.
• Climate Change: Accelerate the transition to renewable energy, invest in fire prevention and mitigation strategies, and adapt infrastructure to withstand extreme weather events.
• Economic Opportunity: Foster a diverse and thriving economy beyond tech, support small businesses, and invest in education and training programs to equip Californians with the skills they need to succeed in the changing job market.

The future of the California Dream is not predetermined. It depends on the choices Californians make today. By embracing innovation, fostering collaboration, and prioritizing both economic prosperity and social equity, California can ensure that the dream remains accessible for generations to come.

This is just the beginning of the conversation. What are your thoughts on the future of California? Share your comments below and let’s keep the dream alive, together.

And now The Point op-ed:

California Crossroads: Dreamin’ Big or Sleepwalkin’ into Disaster?

California, the land of endless sunshine and endless problems. We’re the envy of the nation for our beaches, celebrities, and booming tech industry, but beneath the veneer, cracks are starting to show. Housing prices are astronomical, homelessness is rampant, and wildfires rage with increasing ferocity. Is the California Dream turning into a Californian nightmare?

Don’t get me wrong, California has always been a place to chase big dreams. From the Gold Rush prospectors to the Silicon Valley tech titans, we’ve thrived on a spirit of innovation and audacity. But somewhere along the way, we lost sight of the little guy. The middle class is getting squeezed out, priced out of the housing market and struggling to make ends meet.

Now, let’s not sugarcoat it: California’s got problems. Our elected officials, bless their hearts, sometimes seem more interested in ideological posturing than in solving real problems. We’ve got regulations piled on regulations, strangling businesses and driving up costs. And don’t even get me started on the homeless crisis – it’s a humanitarian disaster unfolding on our streets, and frankly, it’s shameful.

But hold on, here’s the good news: Californians are a resourceful bunch. We’ve faced challenges before, and we’ve always found a way to adapt and overcome. We’re leading the charge on renewable energy, pioneering innovative solutions to tackle climate change, and fostering a diverse and vibrant cultural landscape.

So, what’s the answer? How do we keep the California Dream alive for everyone, not just the privileged few? Here’s my two cents:

• Cut the red tape: Streamline regulations to make it easier for businesses to create jobs and invest in our communities.
• Build, baby, build: Incentivize the construction of affordable housing units and explore creative solutions like tiny homes and co-living arrangements.
• Get tough on homelessness: Implement data-driven solutions that prioritize getting people off the streets and into stable housing, while addressing the root causes of homelessness.
• Embrace innovation: Continue to be a leader in clean energy, sustainable practices, and technological advancements.

California’s future is not set in stone. We can choose to sleepwalk into disaster, or we can roll up our sleeves and get to work. Let’s remember the spirit of audacity that built this state and channel it into finding solutions that benefit all Californians. 

The California Dream may need an upgrade, but it’s not beyond our reach. Let’s get to work, folks, and keep the Golden State shining bright.

Both of them sound right, right?

Obviously, no – there are references to problems but they are all rounded, downplayed, sanded and sifted by Gemini to fit/create the consensus.  And do note how the sustainability, the progressive, diversity and climate change progress attitudes and accomplishments of California needed to be emphasized.

Yup, that sounds like us.

And the consensus is digitized and complete.