And Google's search results are as fake as the images created with their AI generator tool.
The only difference is visual.
As Scott Adams notes, humans are visual creatures. The fakeness just isn't as obvious or compelling in text based search results.
Hopefully this is a Bud Light moment and people will put in some extra effort to finally ditch google products and services. At a minimum you boost your personal privacy!
My father programmed mainframe computers in the '60s and '70s. Dinner table conversation got a little bit techy at times, and some of it rubbed off on me.
My 6th-grade teacher was a young, idealistic sort who wanted to raise awareness of social injustice in her students. Her students were a bunch of 11-year-olds who wanted to do the minimum amount of work possible to get by.
One day she was trying to lead us in a discussion of, I don't remember, inner-city poverty or something, and wasn't having much success. She asked what we thought society should do about this, and we all just shrugged. "Come on," she urged, "you're all going to be adults one day and you'll have to deal with this." One of my classmates raised her hand and said, "By then there will be computers that can solve the problem for us. We'll just do whatever they say." Everyone else nodded in agreement.
My teacher was aghast. How could you surrender your humanity to a soulless machine? My classmates were dumbfounded. If a machine can do the work for us, why not take the easy way out?
After a bit of back and forth, I raised my hand and asked, "Who's going to write the algorithm?" (I told you some of Dad's geek talk rubbed off.)
They had no idea what I was talking about, so I elaborated. "Computers can't do anything on their own; they can only follow instructions in their program. So whoever writes the program will have to know all this poverty stuff already."
Oh, the derision I got for questioning the infallible computer. My teacher sneered at me and said that computers can do anything, they know everything, and their answers are always perfect, so please stop interrupting the discussion.
Oh.
That was more than 50 years ago, and I fear little has changed. Despite the fact that we are surrounded by computers every day, and we all know how dumb they can be, there is still a lot of absolute faith placed in these machines. I saw it on the Internet; it must be true! This is the number Excel gave me; it must be true! And, of course: this was the first search result; it must be the best one! My fear is that we will see whatever AI spits out -- images or text -- and revert to being lazy little 6th-graders again. It came from the computer; it must be right!
And of course, we store all of our factual information on computers now. If AI gets smart enough to start changing primary sources to fit its worldview, we are really screwed.
I asked Gemini to create an image of the 1980 Swedish Olympic Swimming Team — I have a friend who was a member of that team.
Gemini presented me with a picture of eight made-up people two of whom were black. Their racing suits were from various eras. My friend was disgusted.
“Every record has been destroyed or falsified, every book rewritten, every picture has been repainted, every statue and street building has been renamed, every date has been altered. And the process is continuing day by day and minute by minute. History has stopped. Nothing exists except an endless present in which the Party is always right.”
My test question for all AI engines is "What is Instapundit?" Then I argue with it for a while. The argument with Gemini was telling. It called Reynolds a conservative, which is a mistake most other AIs have made. I argued that he is an avowed libertarian. Gemini said he is critical of liberal policies and politicians and that disagreement makes him a conservative.
But Gemini was also a little confused about Reynolds saying: He famously stated, "I'd like to live in a world in which happily married gay people have closets full of assault weapons to protect their pot." This highlights his strong belief in individual liberty beyond typical conservative views.
Therefore, while he might not explicitly say "I'm a conservative," his positions and affiliations lean towards both ideologies, with libertarianism holding a strong influence.
I asked "based on your logic is Reynolds gay?" ...and Gemini lost it in looping illogic and waffled into nothingness. Most AIs identify Instapundit as the individual Glenn Reynolds and not something more complicated and nuanced but in a way that is understandable, especially for a contrived intelligence.
A famous cartoon from ~1980 shows two scientists in lab coats looking at a computer screen. One of them says something like: "If we have artificial intelligence, we also will have artificial stupidity".
Google Search has gotten all but useless in recent years. Ask it a question, and the first 20 entries are “sponsored,” ie, ads. I understand why that’s good for their earnings, but not why it’s useful to me if I’m looking for information, not shopping.
This AI disaster is an unfiltered glimpse into the Big Tech hive mind, just as the scripts of most movies and TV shows are glimpses into the twisted minds of Hollywood screenwriters. And it’s ugly in there. GIGO, increasing woke garbage.
I'm glad Google's initial foray into historical images has been such a joke. As The Professor has noted, when Gemini's all-important first impression is so blatantly untrustworthy it may be quite a while until people believe anything that it produces. "Falsus in uno falsus in omnibus."
The curtain was pulled back and the Normals got to see how ugly and warped Google is. They have destroyed their most valuable possession, Public trust. A picture is worth a thousand words? Then how can you ever trust that Google is giving you honest results? Oh, they will paper over the problem when they roll out Gemini 2.1, but it will have been coded by the same racist engineers, under the direction of the same racist management. Racist, Anti-American Lying firm who is going to do their best again to throw this election to Biden.
The problem is they’ll tweak it to make the bias more subtle. In 10 years the narrative that the US was built by minorities over the obstruction of evil whitey will be pervasive in our schools and youth. They’ll adjust their settings so everything good is portrayed by minorities and everything bad is portrayed by Europeans.
Rather than commenting myself, I am going to share my older son's views on AI. He is a professional programmer--he got into computers as a kid, and started getting paid for the work when he was 20. He's now 49.
"The term "artificial intelligence" misleads people badly. "Machine learning" isn't much better. Neither describes what's actually happening. Which is almost entirely complicated math applied to stupendous amounts of data. Only became possible recently because the computers and storage devices got bigger and faster.
There was a similar hype cycle in the 90's around "expert systems," which was an outgrowth of the Lisp systems and research MIT did dating back to the 60's. As people figure out how much it can and can't do, the hype will die down.
I see some bad things happening with how this stuff is used, but it's mostly because of greed and lack of sense on the part of corporations....Because if you're not an idiot, you realize that selling junk has long-term consequences. Something American middle-management is uniquely terrible at grasping."
The one thing I would add to his view of "American middle-management" is that "upper-level management" doesn't seem to be any better these days.
Analog is, generally, far and away superior to many digital items. Paper ballots vs. digital ballots, for example. An analog dial type speedometer vs. digital number speedometer. In fact, I find the digital control systems in my car to be generally a really bad driving experience compared to my analog driving. And AI as an instruction and learning tool is going to present the same issues vs. humans and written text.
At my VA they encourage disturbed veterans to draw pictures so therapists can diagnose their troubles. Like the revelation of Claudine Gay as the Creature from Harvard's Id, Google's grotesque historical Rorschachs have publicly unmasked the intellectual confusion that underlies the anodyne AI phrases and politically correct posturing of woke firms and educational institutions. Many critics have detected its footprints in our public schools; this is a frightening look at the thing itself — clear, precisely executed illustrations of a crazy, chaotic and disordered mental universe, lying at a lunatic angle to ascertainable fact.
YMMV. Mine did. I asked it to draw a Confederate general. It refused, explaining that normalizing or romanticizing the Confederacy would be harmful. It seems like I should just go to the safest AI, GOODY-2, and call it done. https://www.goody2.ai/
Spot on.
And Google's search results are as fake as the images created with their AI generator tool.
The only difference is visual.
As Scott Adams notes, humans are visual creatures. The fakeness just isn't as obvious or compelling in text based search results.
Hopefully this is a Bud Light moment and people will put in some extra effort to finally ditch google products and services. At a minimum you boost your personal privacy!
This is a bit long, but I promise it's relevant.
My father programmed mainframe computers in the '60s and '70s. Dinner table conversation got a little bit techy at times, and some of it rubbed off on me.
My 6th-grade teacher was a young, idealistic sort who wanted to raise awareness of social injustice in her students. Her students were a bunch of 11-year-olds who wanted to do the minimum amount of work possible to get by.
One day she was trying to lead us in a discussion of, I don't remember, inner-city poverty or something, and wasn't having much success. She asked what we thought society should do about this, and we all just shrugged. "Come on," she urged, "you're all going to be adults one day and you'll have to deal with this." One of my classmates raised her hand and said, "By then there will be computers that can solve the problem for us. We'll just do whatever they say." Everyone else nodded in agreement.
My teacher was aghast. How could you surrender your humanity to a soulless machine? My classmates were dumbfounded. If a machine can do the work for us, why not take the easy way out?
After a bit of back and forth, I raised my hand and asked, "Who's going to write the algorithm?" (I told you some of Dad's geek talk rubbed off.)
They had no idea what I was talking about, so I elaborated. "Computers can't do anything on their own; they can only follow instructions in their program. So whoever writes the program will have to know all this poverty stuff already."
Oh, the derision I got for questioning the infallible computer. My teacher sneered at me and said that computers can do anything, they know everything, and their answers are always perfect, so please stop interrupting the discussion.
Oh.
That was more than 50 years ago, and I fear little has changed. Despite the fact that we are surrounded by computers every day, and we all know how dumb they can be, there is still a lot of absolute faith placed in these machines. I saw it on the Internet; it must be true! This is the number Excel gave me; it must be true! And, of course: this was the first search result; it must be the best one! My fear is that we will see whatever AI spits out -- images or text -- and revert to being lazy little 6th-graders again. It came from the computer; it must be right!
And of course, we store all of our factual information on computers now. If AI gets smart enough to start changing primary sources to fit its worldview, we are really screwed.
I asked Gemini to create an image of the 1980 Swedish Olympic Swimming Team — I have a friend who was a member of that team.
Gemini presented me with a picture of eight made-up people two of whom were black. Their racing suits were from various eras. My friend was disgusted.
“Every record has been destroyed or falsified, every book rewritten, every picture has been repainted, every statue and street building has been renamed, every date has been altered. And the process is continuing day by day and minute by minute. History has stopped. Nothing exists except an endless present in which the Party is always right.”
― George Orwell, 1984
Artificial Intelligence. It's artificial alright.
Blackface. It’s okay when *they* do it.
My test question for all AI engines is "What is Instapundit?" Then I argue with it for a while. The argument with Gemini was telling. It called Reynolds a conservative, which is a mistake most other AIs have made. I argued that he is an avowed libertarian. Gemini said he is critical of liberal policies and politicians and that disagreement makes him a conservative.
But Gemini was also a little confused about Reynolds saying: He famously stated, "I'd like to live in a world in which happily married gay people have closets full of assault weapons to protect their pot." This highlights his strong belief in individual liberty beyond typical conservative views.
Therefore, while he might not explicitly say "I'm a conservative," his positions and affiliations lean towards both ideologies, with libertarianism holding a strong influence.
I asked "based on your logic is Reynolds gay?" ...and Gemini lost it in looping illogic and waffled into nothingness. Most AIs identify Instapundit as the individual Glenn Reynolds and not something more complicated and nuanced but in a way that is understandable, especially for a contrived intelligence.
A famous cartoon from ~1980 shows two scientists in lab coats looking at a computer screen. One of them says something like: "If we have artificial intelligence, we also will have artificial stupidity".
Google Search has gotten all but useless in recent years. Ask it a question, and the first 20 entries are “sponsored,” ie, ads. I understand why that’s good for their earnings, but not why it’s useful to me if I’m looking for information, not shopping.
This AI disaster is an unfiltered glimpse into the Big Tech hive mind, just as the scripts of most movies and TV shows are glimpses into the twisted minds of Hollywood screenwriters. And it’s ugly in there. GIGO, increasing woke garbage.
I'm glad Google's initial foray into historical images has been such a joke. As The Professor has noted, when Gemini's all-important first impression is so blatantly untrustworthy it may be quite a while until people believe anything that it produces. "Falsus in uno falsus in omnibus."
The curtain was pulled back and the Normals got to see how ugly and warped Google is. They have destroyed their most valuable possession, Public trust. A picture is worth a thousand words? Then how can you ever trust that Google is giving you honest results? Oh, they will paper over the problem when they roll out Gemini 2.1, but it will have been coded by the same racist engineers, under the direction of the same racist management. Racist, Anti-American Lying firm who is going to do their best again to throw this election to Biden.
Divest!
Bun it down!
Let no two stones stand together!
Salt the Earth!
Google Delenda Est
The problem is they’ll tweak it to make the bias more subtle. In 10 years the narrative that the US was built by minorities over the obstruction of evil whitey will be pervasive in our schools and youth. They’ll adjust their settings so everything good is portrayed by minorities and everything bad is portrayed by Europeans.
Rather than commenting myself, I am going to share my older son's views on AI. He is a professional programmer--he got into computers as a kid, and started getting paid for the work when he was 20. He's now 49.
"The term "artificial intelligence" misleads people badly. "Machine learning" isn't much better. Neither describes what's actually happening. Which is almost entirely complicated math applied to stupendous amounts of data. Only became possible recently because the computers and storage devices got bigger and faster.
There was a similar hype cycle in the 90's around "expert systems," which was an outgrowth of the Lisp systems and research MIT did dating back to the 60's. As people figure out how much it can and can't do, the hype will die down.
I see some bad things happening with how this stuff is used, but it's mostly because of greed and lack of sense on the part of corporations....Because if you're not an idiot, you realize that selling junk has long-term consequences. Something American middle-management is uniquely terrible at grasping."
The one thing I would add to his view of "American middle-management" is that "upper-level management" doesn't seem to be any better these days.
Analog is, generally, far and away superior to many digital items. Paper ballots vs. digital ballots, for example. An analog dial type speedometer vs. digital number speedometer. In fact, I find the digital control systems in my car to be generally a really bad driving experience compared to my analog driving. And AI as an instruction and learning tool is going to present the same issues vs. humans and written text.
Actually it's AS. Artificial Stupidity.
At my VA they encourage disturbed veterans to draw pictures so therapists can diagnose their troubles. Like the revelation of Claudine Gay as the Creature from Harvard's Id, Google's grotesque historical Rorschachs have publicly unmasked the intellectual confusion that underlies the anodyne AI phrases and politically correct posturing of woke firms and educational institutions. Many critics have detected its footprints in our public schools; this is a frightening look at the thing itself — clear, precisely executed illustrations of a crazy, chaotic and disordered mental universe, lying at a lunatic angle to ascertainable fact.
YMMV. Mine did. I asked it to draw a Confederate general. It refused, explaining that normalizing or romanticizing the Confederacy would be harmful. It seems like I should just go to the safest AI, GOODY-2, and call it done. https://www.goody2.ai/