Just One Big Google Zircon Gemstone for March 5, 2024
March 5, 2024
This essay is the work of a dumb dinobaby. No smart software required.
I have a folder stuffed with Google gems for the week of February 26 to March 1, 12023. I have a write up capturing more Australians stranded by following Google Maps’s representation of a territory, Google’s getting tangled in another publisher lawsuit, Google figuring out how to deliver better search even when the user’s network connection sucks, Google’s firing 43 unionized contractors while in the midst of a legal action, and more.
The brilliant and very nice wizard adds, “Yes, we have created a thing which looks valuable, but it is laboratory-generated. And it is gem and a deeply flawed one, not something we can use to sell advertising yet”. Thanks, MSFT Copilot Bing thing. Good enough and I liked the unasked for ethnic nuance.
But there is just one story: Google nuked billions in market value and created the meme of the week by making many images the heart and soul of diversity. Pundits wanted one half of the Sundar and Prabhakar comedy show yanked off the stage. Check out Stratechery’s view of Google management’s grasp of leading the company in a positive manner in Gemini and Google’s Culture. The screw up was so bad that even the world’s favorite expert in aircraft refurbishment and modern gas-filled airships spoke up. (Yep, that’s the estimable Sergey Brin!)
In the aftermath of a brilliant PR move, CNBC ran a story yesterday that summed up the February 26 to March 1 Google experience. The title was “Google Co-Founder Sergey Brin Says in Rare Public Appearance That Company ‘Definitely Messed Up’ Gemini Image Launch.” What an incisive comment from one of the father of “clever” methods of determining relevance. The article includes this brilliant analysis:
He also commented on the flawed launch last month of Google’s image generator, which the company pulled after users discovered historical inaccuracies and questionable responses. “We definitely messed up on the image generation,” Brin said on Saturday. “I think it was mostly due to just not thorough testing. It definitely, for good reasons, upset a lot of people.”
That’s the Google “gem.” Amazing.
Stephen E Arnold, March 5, 2024
Techno Bashing from Thumb Typers. Give It a Rest, Please
March 5, 2024
This essay is the work of a dumb dinobaby. No smart software required.
Every generation says that the latest cultural and technological advancements make people stupider. Novels were trash, the horseless carriage ruined traveling, radio encouraged wanton behavior, and the list continues. Everything changed with the implementation of television aka the boob tube. Too much television does cause cognitive degradation. In layman’s terms, it means the brain goes into passive functioning rather than actively thinking. It would be almost a Zen moment. Addiction is fun for some.
The introduction of videogames, computers, and mobile devices augmented the decline of brain function. The combination of AI-chatbots and screens, however, might prove to be the ultimate dumbing down of humans. APA PsycNet posted a new study by Umberto León-Domínguez called, “Potential Cognitive Risks Of Generative Transformer-Based AI-Chatbots On Higher Order Executive Thinking.”
Psychologists already discovered that spending too much time on a screen (i.e. playing videogames, watching TV or YouTube, browsing social media, etc.) increases the risk of depression and anxiety. When that is paired with AI-chatbots, or programs designed to replicate the human mind, humans rely on the algorithms to think for them.
León-Domínguez wondered if too much AI-chatbot consumption impaired cognitive development. In his abstract he invented some handy new terms that:
“The “neuronal recycling hypothesis” posits that the brain undergoes structural transformation by incorporating new cultural tools into “neural niches,” consequently altering individual cognition. In the case of technological tools, it has been established that they reduce the cognitive demand needed to solve tasks through a process called “cognitive offloading.” Cognitive offloading”perfectly describes younger generations and screen addicts. “Cultural tools into neural niches” also respects how older crowds view new-fangled technology, coupled with how different parts of the brain are affected with technology advancements. The modern human brain works differently from a human brain in the 18th-century or two thousand years ago.
He found:
“The pervasive use of AI chatbots may impair the efficiency of higher cognitive functions, such as problem-solving. Importance: Anticipating AI chatbots’ impact on human cognition enables the development of interventions to counteract potential negative effects. Next Steps: Design and execute experimental studies investigating the positive and negative effects of AI chatbots on human cognition.”
Are we doomed? No. Do we need to find ways to counteract stupidity? Yes. Do we know how it will be done? No.
Isn’t tech fun?
Whitney Grace, March 6, 2024
SearXNG: A New Metasearch Engine
March 4, 2024
This essay is the work of a dumb dinobaby. No smart software required.
Internet browsers and search engines are two of the top applications used on computers. Search engine giants like Bing and Google don’t respect users’ privacy and they track everything. They create individual user profiles then sell and use the information for targeted ads. The search engines also demote controversial information and return biased search results. On his blog, FlareXes shares a solution that protects privacy and encompasses metasearch: “Build Your Own Private Search Engine With SearXNG.”
SearXNG is an open source, customizable metasearch engine that returns search results from multiple sources and respects privacy. It was originally built off another open source project SearX. SearXNG has an extremely functional user interface. It also aggregates information from over seventy search engines, including DuckDuckGo, Brave Search, Bing, and Google.
The best thing about SearXNG is protecting user privacy: But perhaps the best thing about SearXNG is its commitment to user privacy. Unlike some search engines, SearXNG doesn’t track users or generate personalized profiles, and it never shares any information with third parties.”
Because SearXNG is a metasearch engine, it supports organic search results. This allows users to review information that would otherwise go unnoticed. That doesn’t mean the returns will allegedly be unbiased. The idea is that SearXNG returns better results than a revenue juggernaut:
SearXNG aggregates data from different search engines that doesn’t mean this could be biased. There is no way for Google to create a profile about you if you’re using SearXNG. Instead, you get high-quality results like Google or Bing. SearXNG also randomizes the results so no SEO or top-ranking will not gonna work. You can also enable independent search engines like Brave Search, Mojeek etc.”
If you want a search engine that doesn’t collect your personal data and has betters search results, warrants a test drive. The installation may require some tech fiddling.
Whitney Grace, March 4, 2024
Synthetic Data: From Science Fiction to Functional Circumscription
March 4, 2024
This essay is the work of a dumb humanoid. No smart software required.
Synthetic data are information produced by algorithms, not by real-world events. It’s created using real-world data and numerical recipes. The appeal is that it is easier than collecting real life information, cheaper than dealing with data from real life, and faster than fooling around with surveys, monitoring devices, and law suits. In theory, synthetic data is one promising way of skirting the expense of getting humans involved.
“What Is [a] Synthetic Sample – And Is It All It’s Cracked Up to Be?” tackles the subject of a synthetic sample, a topic which is one slice of the synthetic data universe. The article seeks “to uncover the truth behind artificially created qualitative and quantitative market research data.” I am going to avoid the question, “Is synthetic data useful?” because the answer is, “Yes.” Bean counters and those looking to find a way out of the pickle barrel filled with expensive brine are going to chase after the magic of algorithms producing data to do some machine learning magic.
In certain situations, fake flowers are super. Other times, the faux blooms are just creepy. Thanks, MSFT Copilot Bing thing. Good enough.
Are synthetic data better than real world data? The answer from my vantage point is, “It depends.” Fancy math can prove that for some use cases, synthetic data are “good enough”; that is, the data produce results close enough to what a “real” data set provides. Therefore, just use synthetic data. But for other applications, synthetic data might throw some sand in the well-oiled marketing collateral describing the wonders of synthetic data. (Some university research labs are quite skilled in PR speak, but the reality of their methods may not line up with the PowerPoints used to raise venture capital.)
This essay discusses a research project to figure out if a synthetic sample works or in my lingo if the synthetic sample is good enough. The idea is that as long as the synthetic data is within a specified error range, the synthetic sample can be used and may produce “reliable” or useful results. (At least one hopes this is the case.)
I want to focus on one portion of the cited article and invite you to read the complete Kantar explanation.
Here’s the passage which snagged my attention:
… right now, synthetic sample currently has biases, lacks variation and nuance in both qual and quant analysis. On its own, as it stands, it’s just not good enough to use as a supplement for human sample. And there are other issues to consider. For instance, it matters what subject is being discussed. General political orientation could be easy for a large language model (LLM), but the trial of a new product is hard. And fundamentally, it will always be sensitive to its training data – something entirely new that is not part of its training will be off-limits. And the nature of questioning matters – a highly ’specific’ question that might require proprietary data or modelling (e.g., volume or revenue for a particular product in response to a price change) might elicit a poor-quality response, while a response to a general attitude or broad trend might be more acceptable.
These sentences present several thorny problems is academic speak. Let’s look at them in the vernacular of rural Kentucky where I live.
First, we have the issue of bias. Training data can be unintentionally or intentionally biased. Sample radical trucker posts on Telegram, and use those messages to train a model like Reor. That output is going to express views that some people might find unpalatable. Therefore, building a synthetic data recipe which includes this type of Telegram content is going to be oriented toward truck driver views. That’s good and bad.
Second, a synthetic sample may require mixing data from a “real” sample. That’s a common sense approach which reduces some costs. But will the outputs be good enough. The question then becomes, “Good enough for what applications?” Big, general questions about how a topic is presented might be close enough for horseshoes. Other topics like those focusing on dealing with a specific technical issue might warrant more caution or outright avoidance of synthetic data. Do you want your child or wife to die because the synthetic data about a treatment regimen was close enough for horseshoes. But in today’s medical structure, that may be what the future holds.
Third, many years ago, one of the early “smart” software companies was Autonomy, founded by Mike Lynch. In the 1990s, Bayesian methods were known but some — believe it or not — were classified and, thus, not widely known. Autonomy packed up some smart software in the Autonomy black box. Users of this system learned that the smart software had to be retrained because new terms and novel ideas not in the original training set were not findable by the neuro linguistic program’s engine. Yikes, retraining requires human content curation of data sets, time to retrain the system, and the expense of redeploying the brains of the black boxes. Clients did not like this and some, to be frank, did not understand why a product did not work like an MG sports car. Synthetic data has to be trained to “know” about new terms and avid the “certain blindness” probability based systems possess.
Fourth, the topic of “proprietary data modeling” means big bucks. The idea behind synthetic data is that it is cheaper. Building proprietary training data and keeping it current is expensive. Is it better? Yeah, maybe. Is it faster? Probably not when humans are doing the curation, cleaning, verifying, and training.
The write up states:
But it’s likely that blended models (human supplemented by synthetic sample) will become more common as LLMs get even more powerful – especially as models are finetuned on proprietary datasets.
Net net: Synthetic data warrants monitoring. Some may want to invest in synthetic data set companies like Kantar, for instance. I am a dinobaby, and I like the old-fashioned Stone Age approach to data. The fancy math embodies sufficient risk for me. Why increase risk? Remember my reference to a dead loved one? That type of risk.
Stephen E Arnold, March 4, 2023
Technology Becomes Detroit
March 4, 2024
This essay is the work of a dumb dinobaby. No smart software required.
Have you ever heard of technical debt? Technical debt is when an IT team prioritize speedy delivery of a product over creating a feasible, quality product. Technology history is full of technical debt. Some of the more famous cases are the E.T. videogame for the Atari, Windows Vista, and the Samsung Galaxy Gear. Technical debt is an ongoing issue for IT departments and tech companies. It’s apparently getting worse. ITPro details the current problems with technical debt in, “IT Leaders Need To Accept They’ll Never Escape Technical Debt, But That Doesn’t Mean They Should Down Tools.”
Gordon Haff is a senior leader at Red Hat and a technology evangelist. Haff told ITPro that tech experts will continue to remain hindered as they continue to deal with technical debt and shill shortages. Tech experts want to advance their field with transformative projects but they’re held back by the same aforementioned issues. Haff stressed that as soon as one project is complete, tech experts build the next project on existing architecture. It creates a technical debt infrastructure.
Haff provided an example using a band-aid metaphor:
“Haff pointed toward application modernization as a prime example of this rinse and repeat trend. Many enterprises, he said, deliberately choose to not tinker with certain applications due to the fact they still worked nominally.
Fast forward several years later, these applications are overhauled and modernized, then are left to their own devices – to some extent – and reassessed during the next transformation cycle.
‘If you go back 10 years, we had this sort of bimodal IT, or fast-slow IT, that was kind of the thing,” he explained. “The idea was ‘we’ll leave that old stuff, we’ll shove that off into the corner and not worry about it’ and the cool kids can work on all this greenfield, often new customer-facing applications.
‘But by and large, it’s then a case of ‘oh we actually need to deal with this core business stuff’ and these older applications.’”
Haff suggests that IT experts shouldn’t approach their work with a “one and done” mindset. They should realize their work is constantly evolving. These should be aware of how to go with the flow and program legacy systems that don’t transform into large messes. There’s a reason videogame companies have beta tests, restaurants have soft openings, and musicals have previews. They test things to deliver quality products. Technical debt leads to technical rot.
Whitney Grace, March 4, 2024
Forget the Words. Do Short-Form Video by Hiring a PR Professional
March 1, 2024
This essay is the work of a dumb humanoid. No smart software required.
I think “Everyone’s a Sellout Now” is about 4,000 words. The main idea is that traditional publishing is roached. Artists and writers must learn to do video editing or have enough of mommy and daddy’s money to pay someone to promote the creator’s output. The essay is well written; however, I am not sure it conveys a TikTok fact unknown or hiding in the world of BlueSky-type services.
This bright young student should have used a ChatGPT-type service. Thanks, MSFT Copilot. At least you are outputting which is more than I can say for your fierce but lagging competitor.
I noted this passage:
Because self-promotion sucks.
I think I agree, but why not hire an “output handler.” The OH does the PR.
Here’s another quote to note:
The problem is that America more or less runs on the concept of selling out.
Is there a fix for the gasoline of America? Yes. The essay asserts:
author-content creators succeed by making the visually uninteresting labor of typing on a laptop worthwhile to watch.
The essay concludes with this less-than-uplifting comment:
To achieve the current iteration of the American dream, you’ve got to shout into the digital void and tell everyone how great you are. All that matters is how many people believe you.
Downer? Yes, and what makes it fascinating is that the author gets paid for writing. I think this is a “real job.”
Several observations:
- I think smart software is going to do more than write wacko stuff for SmartNews-type publications.
- Readers of “downer” essays are likely to go more “down”; that is, become less positive and increasingly antagonistic to what makes the US of A tick
- The essay delivers the news about the importance of TikTok without pointing out that the service is China-affiliated and provides content not permitted for consumption in China.
Net net: Hire a gig worker to do the OH. Pay for PR. Quit complaining or complain in fewer words.
PS. The categorical affirmative of “everyone” is disproved with a single example. As I have pointed out in an essay about a grousing Xoogler, I operate differently. Therefore, the everyone is like fuzzy antecedents. Sloppy.
Stephen E Arnold, March 1, 2024
Bad News Delivered via Math
March 1, 2024
This essay is the work of a dumb humanoid. No smart software required.
I am not going to kid myself. Few people will read “Hallucination is Inevitable: An Innate Limitation of Large Language Models” with their morning donut and cold brew coffee. Even fewer will believe what the three amigos of smart software at the National University of Singapore explain in their ArXiv paper. Hard on the heels of Sam AI-Man’s ChatGPT mastering Spanglish, the financial payoffs are just too massive to pay much attention to wonky outputs from smart software. Hey, use these methods in Excel and exclaim, “This works really great.” I would suggest that the AI buggy drivers slow the Kremser down.
The killer corollary. Source: Hallucination is Inevitable: An Innate Limitation of Large Language Models.
The paper explains that large language models will be reliably incorrect. The paper includes some fancy and not so fancy math to make this assertion clear. Here’s what the authors present as their plain English explanation. (Hold on. I will give the dinobaby translation in a moment.)
Hallucination has been widely recognized to be a significant drawback for large language models (LLMs). There have been many works that attempt to reduce the extent of hallucination. These efforts have mostly been empirical so far, which cannot answer the fundamental question whether it can be completely eliminated. In this paper, we formalize the problem and show that it is impossible to eliminate hallucination in LLMs. Specifically, we define a formal world where hallucination is defined as inconsistencies between a computable LLM and a computable ground truth function. By employing results from learning theory, we show that LLMs cannot learn all of the computable functions and will therefore always hallucinate. Since the formal world is a part of the real world which is much more complicated, hallucinations are also inevitable for real world LLMs. Furthermore, for real world LLMs constrained by provable time complexity, we describe the hallucination-prone tasks and empirically validate our claims. Finally, using the formal world framework, we discuss the possible mechanisms and efficacies of existing hallucination mitigators as well as the practical implications on the safe deployment of LLMs.
Here’s my take:
- The map is not the territory. LLMs are a map. The territory is the human utterances. One is small and striving. The territory is what is.
- Fixing the problem requires some as yet worked out fancier math. When will that happen? Probably never because of no set can contain itself as an element.
- “Good enough” may indeed by acceptable for some applications, just not “all” applications. Because “all” is a slippery fish when it comes to models and training data. Are you really sure you have accounted for all errors, variables, and data? Yes is easy to say; it is probably tough to deliver.
Net net: The bad news is that smart software is now the next big thing. Math is not of too much interest, which is a bit of a problem in my opinion.
Stephen E Arnold, March 1, 2024
Student Surveillance: It Is a Thing
March 1, 2024
This essay is the work of a dumb dinobaby. No smart software required.
Once mobile phones were designed with cameras, all technology was equipped with one. Installing cameras and recording devices is SOP now, but facial recognition technology will soon become as common unless privacy advocates have their way. Students at the University of Waterloo were upset to learn that vending machines on their campus were programmed with the controversial technology. The Kitchener explores how the scandal started in: “ ‘Facial Recognition’ Error Message On Vending Machine Sparks Concern At University Of Waterloo.”
A series of smart vending machines decorated with M&M graphics and dispense candy were located throughout the Waterloo campus. They raised privacy concerns when a student noticed an error message about the facial recognition application on one machine. The machines were then removed from campus. Until they were removed, word spread quickly and students covered a hole believed to hold a camera.
Students believed that vending machines didn’t need to have facial recognition applications. They also wondered if there were more places on campus where they were being monitored with similar technology.
The vending machines are owned by MARS, an international candy company, and manufactured by Invenda. The MARS company didn’t respond to queries but Invenda shared more information about the facial recognition application:
“Invenda also did not respond to CTV’s requests for comment but told Stanley in an email ‘the demographic detection software integrated into the smart vending machine operates entirely locally.’ ‘It does not engage in storage, communication, or transmission of any imagery or personally identifiable information,’ it continued.
According to Invenda’s website, the Smart Vending Machines can detect the presence of a person, their estimated age and gender. The website said the ‘software conducts local processing of digital image maps derived from the USB optical sensor in real-time, without storing such data on permanent memory mediums or transmitting it over the Internet to the Cloud.’”
Invenda also said the software is compliant with the European Union privacy General Data Protection Regulation but that doesn’t mean it is legal in Canada. The University of Waterloo has asked that the vending machines be removed from campus.
Net net: Cameras will proliferate and have smart software. Just a reminder.
Whitney Grace, March 1, 2024
Open Source: Free, Easy, and Fast Sort Of
February 29, 2024
This essay is the work of a dumb dinobaby. No smart software required.
Not long ago, I spoke with an open source cheerleader. The pros outweighed the cons from this technologist’s point of view. (I would like to ID the individual, but I try to avoid having legal eagles claw their way into my modest nest in rural Kentucky. Just plug in “John Wizard Doe”, a high profile entrepreneur and graduate of a big time engineering school.)
I think going up suggests a problem.
Here are highlights of my notes about the upside of open source:
- Many smart people eyeball the code and problems are spotted and fixed
- Fixes get made and deployed more rapidly than commercial software which of works on an longer “fix” cycle
- Dead end software can be given new kidneys or maybe a heart with a fork
- For most use cases, the software is free or cheaper than commercial products
- New functions become available; some of which fuel new product opportunities.
There may be a few others, but let’s look at a downside few open source cheerleaders want to talk about. I don’t want to counter the widely held belief that “many smart people eyeball the code.” The method is grab and go. The speed angle is relative. Reviving open source again and again is quite useful; bad actors do this. Most people just recycle. The “free” angle is a big deal. Everyone like “free” because why not? New functions become available so new markets are created. Perhaps. But in the cyber crime space, innovation boils down to finding a mistake that can be exploited with good enough open source components, often with some mileage on their chassis.
But the one point open source champions crank back on the rah rah output. “Over 100,000 Infected Repos Found on GitHub.” I want to point out that GitHub is a Microsoft, the all-time champion in security, owns GitHub. If you think about Microsoft and security too much, you may come away confused. I know I do. I also get a headache.
This “Infected Repos” API IRO article asserts:
Our security research and data science teams detected a resurgence of a malicious repo confusion campaign that began mid-last year, this time on a much larger scale. The attack impacts more than 100,000 GitHub repositories (and presumably millions) when unsuspecting developers use repositories that resemble known and trusted ones but are, in fact, infected with malicious code.
The write up provides excellent information about how the bad repos create problems and provides a recipe for do this type of malware distribution yourself. (As you know, I am not too keen on having certain information with helpful detail easily available, but I am a dinobaby, and dinobabies have crazy ideas.)
If we confine our thinking to the open source champion’s five benefits, I think security issues may be more important in some use cases.The better question is, “Why don’t open source supporters like Microsoft and the person with whom I spoke want to talk about open source security?” My view is that:
- Security is an after thought or a never thought facet of open source software
- Making money is Job #1, so free trumps spending money to make sure the open source software is secure
- Open source appeals to some venture capitalists. Why? RedHat, Elastic, and a handful of other “open source plays”.
Net net: Just visualize a future in which smart software ingests poisoned code, and programmers who rely on smart software to make them a 10X engineer. Does that create a bit of a problem? Of course not. Microsoft is the security champ, and GitHub is Microsoft.
Stephen E Arnold, February 29, 2024
Surprise! Smart Software and Medical Outputs May Kill You
February 29, 2024
This essay is the work of a dumb humanoid. No smart software required.
Have you been inhaling AI hype today? Exhale slowly, then read “Generating Medical Errors: GenAI and Erroneous Medical References,” produced by the esteemed university with a reputation for shaking the AI cucarachas and singing loudly “Ai, Ai, Yi.” The write up is an output of the non-plagiarizing professionals in the Human Centered Artificial Intelligence unit.
The researchers report states:
…Large language models used widely for medical assessments cannot back up claims.
Here’s what the HAI blog post states:
we develop an approach to verify how well LLMs are able to cite medical references and whether these references actually support the claims generated by the models. The short answer: poorly. For the most advanced model (GPT-4 with retrieval augmented generation), 30% of individual statements are unsupported and nearly half of its responses are not fully supported.
Okay, poorly. The disconnect is that the output sounds good, but the information is distorted, off base, or possibly inappropriate.
What I found interesting is a stack ranking of widely used AI “systems.” Here’s the chart from the HAI article:
The least “poor” are the Sam AI-Man systems. In the middle is the Anthropic outfit. Bringing up the rear is the French “small” LLM Mistral system. And guess which system is dead last in this Stanford report?
Give up?
The Google. And not just the Google. The laggard is the Gemini system which was Bard, a smart software which rolled out after the Softies caught the Google by surprise about 14 months ago. Last in URL validity, last in statement level support, and last in response level support.
The good news is that most research studies are non reproducible or, like the former president of Stanford’s work, fabricated. As a result, I think these assertions will be easy for an art history major working in Google’s PR confection machine will bat them away like annoying flies in Canberra, Australia.
But last from researchers at the estimable institution where Google, Snorkel and other wonderful services were invented? That’s a surprise like the medical information which might have unexpected consequences for Aunt Mille or Uncle Fred.
Stephen E Arnold, February 29, 2024
-
- Subscribe to Beyond Search
Feature archive
News archive
-
