More on the AI Betamax Versus VHS Dust Up
February 2, 2023
“24 Seriously Embarrassing Hours for AI” gathers four smart software stumbles. The examples are highly suggestive that some butchers have been putting their fingers on the scales. The examples include the stage set approach to Tesla’s self driving and OpenAI’s reliance on humans to beaver away out of sight to make outputs better. In general I agree with the points in the write up.
However, there is one statement which attracted my yellow high light pen like a sci-fi movie tractor beam. Here it is:
Sometimes the slower road is the better road.
It may be that the AI TGV has already left the station and is hurtling down the rails from Paris to Nimes. Microsoft announced that the lovable Teams video chat and Swiss Army knife of widgets will be helping users lickity split. Other infusions are almost certain to follow. Even airlines are thinking smart software. Airlines! These outfits lose luggage with bar codes. Perhaps AI will help, but I remain skeptical. How does one lose a bag with a bar code in our post 9/11 world?
The challenge for Google, Facebook (which wants to be a leader in AI), and the other organizations betting their investors’ money on AI going to take a “slower road”?
My TGV high speed train reference is not poetical; it is a reflection of the momentum of information. The OpenAI machine — with or without legerdemain — is rolling along. OpenAI has momentum. With foresight or dumb luck, Microsoft is riding along.
The “slower road” echoes Google’s conservative approach. Remember that Google sacrificed credibility in AI with the Dr. Timnit Gebru affair. Like a jockey with a high value horse, the beast is now carrying lead pads. Combine that with bureaucratic bloat and concern for ad revenues, I am not sure Google and some other outfits can become the high twitch muscled creature needed to cope with market momentum.
Betamax was better. Well, it did not dominate the market. VHS was pushed into the ditch, but that required time and technological innovation. The AI race is not over but the “slow” angle is late from the gate.
Stephen E Arnold, February 2, 2023
Does Google Have the Sony Betamax of Smart Software?
January 30, 2023
Does Google have the Sony Betamax of smart software? If you cannot answer this question as well as ChatGPT, you can take a look at “VHS or Beta? A Look Back at Betamax, and How Sony Lost the VCR Format War to VHS Recorders.” Boiling down the problem Sony faced, let me suggest better did not win. Maybe adult content outfits tipped the scales? Maybe not? The best technology does not automatically dominate the market.
Flash forward from the anguish of Sony in the 1970s and the even more excruciating early 1980s to today. Facebook dismisses ChatGPT as not too sophisticated. I heard one of the big wizards at the Zuckbook say this to a Sillycon Alley journalist on a podcast called Big Technology. The name says it all. Big technology, just not great technology. That’s what the Zuckbooker suggested everyone’s favorite social media company has.
The Google has emitted a number of marketing statements about more than a dozen amazing smart software apps. These, please, note, will be forthcoming. The most recent application of the Google’s advanced, protein folding, Go winning system is explained in words—presumably output by a real journalist—in “Google AI Can Create Music in Any Genre from a Text Description.” One can visualize the three exclamation points that a human wanted to insert in this headline. Amazing, right. That too is forthcoming. The article quickly asserts something that could have been crafted by one of Googzilla’s non-terminated executives believes:
MusicLM is surprisingly talented.
The GOOG has talent for sure.
What the Google does not have is the momentum of consumer craziness. Whether it the buzz among some high school and college students that ChatGPT can write or help write term papers or the in-touch outfit Buzzfeed which will use ChatGPT to create listicles — the indomitable Alphabet is not in the information flow.
But the Google technology is better. That sounds like a statement I heard from a former wizard at RCA who was interviewing for a job at the blue chip consulting firm for which I worked when I was a wee lad. That fellow invented some type of disc storage system, maybe a laser-centric system. I don’t know. His statement still resonates with me today:
The Sony technology was better.
The flaw is that the better technology can win. The inventors of the better technology or the cobblers who glue together other innovations to create a “better” technology never give up their convictions. How can a low resolution, cheaper recording solution win? The champions of Sony’s technology complained about fairness a superior resolution for the recorded information.
I jotted down this morning (January28, 2023), why Googzilla may be facing, like the Zuckbook, a Sony Betamax moment:
- The demonstrations of the excellence of the Google smart capabilities are esoteric and mean essentially zero outside of the Ivory Tower worlds of specialists. Yes, I am including the fans of Go and whatever other game DeepMind can win. Fan frenzy is not broad consumer uptake and excitement.
- Applications which ordinary Google search users can examine are essentially vaporware. The Dall-E and ChatGPT apps are coming fast and furious. I saw a database of AI apps based on these here-and-now systems, and I had no idea so many clever people were embracing the meh-approach of OpenAI. “Meh,” obviously may not square with what consumers perceive or experience. Remember those baffled professors or the Luddite lawyers who find smart software a bit of a threat.
- OpenAI has hit a marketing home run. Forget the Sillycon Alley journalists. Think about the buzz among the artists about their potential customers typing into a search box and getting an okay image. Take a look at Googzilla trying to comprehend the Betamax device.
Toss in the fact that Google’s ad business is going to have some opportunities to explain why owning the bar, the stuff on the shelves, the real estate, and the payment system is a net gain for humanity. Yeah, that will be a slam dunk, won’t it?
Perhaps more significantly, in the post-Covid crazy world in which those who use computers reside, the ChatGPT and OpenAI have caught a big wave. That wave can swamp some very sophisticated, cutting edge boats in a short time.
Here’s a question for you (the last one in this essay I promise): Can the Google swim?
Stephen E Arnold, January 30, 2023
Responding to the PR Buzz about ChatGPT: A Tale of Two Techies
January 24, 2023
One has to be impressed with the PR hype about ChatGPT. One can find tip sheets for queries (yes, queries matter… a lot), ideas for new start ups, and Sillycon Valley pundits yammering on podcasts. At an Information Industry Association meting in Boston, Massachusetts, a person whom I think was called Marvin or Martin Wein-something made an impassioned statement about the importance of artificial intelligence. I recall his saying, “It is happening. Now.”
Marvin or Martin made that statement which still sticks in my mind in 1982 or so. That works out to 40 years ago.
What strikes me this morning is the difference between the response of Facebook and Google. This is a Tale of Two Techies.
In the case of Google, it is Red Alert time. The fear is palpable among the senior managers. How do I know? I read “Google Founders Return As ChatGPT Threatens Search Business.” I could trot out some parallels between Google management’s fear and the royals threatened by riff raff. Make no mistake. The Googlers have quantum supremacy and the DeepMind protein and game playing machine. I recall reading or being told that Google has more than 20 applications that will be available… soon. (Wasn’t that type of announcement once called vaporware?) The key point is that the Googlers are frightened, and like Disney, have had to call the team of Brin and Page to revivify the thinking about the threat to the search business. I want to remind you that the search business was inspired by Yahoo’s Overture approach. Google settled some litigation and the rest is history. Google became an alleged monopoly and Yahoo devolved into a spammy email service.
And what about Facebook? I noted this article: “ChatGPT Is Not Particularly Innovative and Nothing Revolutionary, Says Meta’s Chief AI Scientist.” The write up explains that Meta’s stance with regard to the vibe machine ChatGPT is “meh.” I think Meta or the Zuckbook does care, but Meta has a number of other issues to which the proud firm must respond. Smart software that seems to be a Swiss Army knife of solutions is “nothing revolutionary.” Okay.
Let’s imagine we are in college in one of those miserable required courses in literature. Our assignment is to analyze the confection called the Tale of Two Techies. What’s the metaphorical pivot for this soap opera?
Here’s my take:
- Meta is either too embarrassed, too confused, or too overwhelmed with on-going legal hassles to worry too much about ChatGPT. Putting on the “meh” is good enough. The company seems to be saying, “We don’t care too much… at least in public.”
- Google is running around with its hair on fire. The senior management team’s calling on the dynamic duo to save the day is indicative of the mental short circuits the company exhibits.
Net net: Good, bad, or indifferent ChatGPT makes clear the lack of what one might call big time management thinking. Is this new? Sadly, no.
Stephen E Arnold, January 24, 2023
Are Facebook and Google Monopolies: Nope, Shrinking Share of Online Ads. Proof!
December 29, 2022
I read an interesting article, but I have my doubts about the numbers. The story is from one of the “last person standing” in the Silicon Valley real news datasphere. In the last month or so, the tone of write ups about two of America’s most lovable and well managed companies has turned south, well, maybe south by southwest.
“Share of US Digital Ad Spend, by Company Type” reports:
Google and Meta will together capture 48.4% of all U.S. digital ad revenue this year (28.8% for Google and 19.6% for Meta), down from 54.7% at their peak in 2017 (34.7% for Google and 20.0% for Meta), per data from Insider Intelligence.
And what about the lovable Bezos bulldozer driven pedal to the metal by Andy Jassy? The article states:
- By far, the biggest threat to their collective ad dominance is Amazon, which has grown its ad business to over $30 billion dollars annually.
- By 2024, Amazon is expected to capture 12.7% of all U.S. digital ad dollars, while Meta is expected to capture 17.9%.
TikTok is no big whoop. I suppose that’s why the tech giants are becoming pretzels in their effort create short form content.
Several observations:
- I am not sure how these data were gathered nor the methods used to present such remarkable precision as 54.7 percent in a prediction is an indication that someone did not pay attention in Statistics 101
- Amazon’s ad data are more interesting when the slope between the firm’s ad revenue in 2018 is plotted against Amazon’s ad revenue in 2021. That a slope!
- Blowing off TikTok is problematic. Does the data consider influencers who accept some type of compensation in return for merchandise, trips, or some other fungible asset like a super duper hair curling device?
To sum up: I am not prepared to label those wonderful wizards at Facebook and Google as crew on a doomed steamship named MY Failure.
Stephen E Arnold, December 2022
Ka-Ching, Ka-Ching: The Unforgettable Tune Haunting Meta
December 26, 2022
What’s the sound of the European cash registers? Ka-ching, ka-ching. The melody of money.
I spotted two allegedly true real news stories this morning that sound like previous write ups about the Zuckbook, oops, Facebook, oh, darn, Meta.
The first is “Meta Settles Cambridge Analytica Class-Action Lawsuit for $725 Million.” The story reports:
Parent company Meta has agreed to pay $725 million to settle a long-running class-action lawsuit accusing Facebook of allowing Cambridge Analytica and other third parties to access user’s private information…
Cambridge Analytica. Four years ago. Links to possible intelligence activities in a certain special operation favoring nation state. Lies. What’s not to love? The is writing a check for $725 million. In terms of fellow traveler Twitter that’s peanuts. No big deal. Guilty? “Senator, thank you for that question….”
The second write up is “EU Tells Meta That Facebook Marketplace Breaches Anti-Trust Rules.” This write up explains:
In yet another demonstration of the new aggressive stance of European regulators, the EU has lodged a notice accusing Meta of ‘abusive practices’. The allegation relates to the fact that Facebook Marketplace currently ties its online classified ads service to the Facebook social network. The European Commission alleges that the company imposes unfair trading conditions on rivals to its Facebook Marketplace, in order to gain a commercial advantage.
How disrespectful! The Zuck has worked to bring people together. The Marketplace does that. I live in rural Kentucky. One of my neighbors was killed meeting a person to purchase a mobile phone. The connection making mechanism seems to have been a certain social network. It is clear to me that I don’t understand the bridge building, good vibe generating mechanisms of social media. That’s okay. I am a dinobaby.
Let’s get back to the music of ka-ching, ka-ching.
I have some expectation that the European Union will be ringing its cash registers in 2023. Some US high-technology companies output cash when the tune ka-ching, ka-ching sounds. That may be a conditioned reflex similar to Ivan Pavlov’s demonstration of conditioning. Yes, ka-ching may be a suitable substitute for dog treats. Woof.
In my opinion, the ka-ching, ka-ching thing suggests:
- A pattern of behavior that flaunts expected business practices
- Management attitudes that encourage behavior some lawmakers consider illegal
- An increasing perception that some of the societal pressure points are becoming more sensitive because of certain social media practices.
Fair or unfair? True or false?
Well, just listen: Ka-ching, ka-ching.
Stephen E Arnold, December 26, 2022
Meta: Big Numbers, Bigger Problem
December 22, 2022
The European Union has been fiddling around with modest fines on outfits like Meta (the Zuckbook to the Arnold IT research team). Now the scale of fines is ratcheting up. “Meta Faces $1.6bn Lawsuit over Facebook Posts Inciting Violence in Tigray War” reports that Kenya is pegging the fine in nine figure territory. Will the Zuckbook write a check for more than $1 billion. Probably not.
The write up states as real news:
The lawsuit, filed in the high court of Kenya, where Meta’s sub-Saharan African operations are based, alleges that Facebook’s recommendations systems amplified hateful and violent posts in the context of the war in northern Ethiopia, which raged for two years until a ceasefire was agreed in early November. The lawsuit seeks the creation of a $1.6bn (£1.3bn) fund for victims of hate speech. One of the petitioners said his father, an Ethiopian academic, was targeted with racist messages before his murder in November 2021, and that Facebook did not remove the posts despite complaints.
What’s interesting is that the lawsuit puts a price penalty for the alleged action or inaction of the Zuckbook. Thus, when harm to individuals can be linked to an online action, the Kenya lawsuit means that other governments may set similar targets.
Net net: It may be expensive to implement the Silicon Valley, high technology “we want to bring people together and do good” under the umbrella of move fast and break things. The fines could, despite the Zuckbook’s revenue prowess, break the bank.
How long has the Zuckbook been using its methods for its advantage? A decade or more? Kenya may be taking steps to make the Zuck wish his company could go back in time and approach the company’s behavior in a slightly different way. If the Zuckbook wins, the Zuckbook may go full speed ahead and become even more frisky: TikTok-type videos on steroids, amped up data collection, and creating a super app designed to make bad actors drool. Disappearing messages. What’s not to like?
Stephen E Arnold, December 22, 2022
The Zuck, Personalized Advertising, and the European Data Protection Board Battle Royale 2023
December 13, 2022
I read “EDPB Adopts Art. 65 Dispute Resolution Binding Decisions Regarding Facebook, Instagram and WhatsApp”. The less-then-exciting prose makes clear that the pesky EU and its GDPR ideas are not going away. The official document, dated December 6, 2022, stated:
The Facebook and Instagram draft decisions concern, in particular, the lawfulness and transparency of processing for behavioural advertising. The WhatsApp draft decision concerns notably the lawfulness of processing for the purpose of the improvement of services. Several SAs issued objections on the draft decisions prepared by the Irish SA concerning, among others, the legal basis for processing (Art. 6 GDPR), data protection principles (Art. 5 GDPR), and the use of corrective measures including fines.
After a few more committee meetings, more information will be posted.
This seems pretty innocuous. Another EU regulation, delays, discussions, and inevitable litigation.
Nevertheless, several observations appear to be warranted by your trusty observer in rural Kentucky:
- Will the EU and its state entities levy fines? My hunch is, “Ka-ching” sound a number of times. Go where the money is before the money runs out.
- Will the personalized ad contagion spread to other US outfits? My initial reaction is, “Not even a China-style Covid lockdown can prevent the problem from spreading, morphing, and befuddling some legal eagles.”
- Will the personalized advertisers change? My instinct is that there will be some change. But it will be inspired by Google’s attempt to deal with its tracking methods.
- Will the issue penetrate the hermetically sealed walls of the Apple spaceship? Let me go out on a limb and suggest, “Yep, Level Four containment will be breached.”
- What does consenting to terms and conditions for a service mean? Here’s my take: “Grounds for legal action because… consumers.”
Stephen E Arnold, December 13, 2022
Zucky, You Get a Bad Grade
December 8, 2022
In effort to expand past its Facebook roots, Meta is venturing in multiple directions. We suspect executives hoped its seemingly noble AI project would do better than its floundering VR initiative. Alas, CNet reveals, “Meta Trained an AI on 48M Science Papers. It Was Shut Down After 2 Days.” Well that was fast. Reporter Jackson Ryan explains:
“The tool is pitched as a kind of evolution of the search engine but specifically for scientific literature. Upon Galactica’s launch, the Meta AI team said it can summarize areas of research, solve math problems and write scientific code. At first, it seems like a clever way to synthesize and disseminate scientific knowledge. Right now, if you wanted to understand the latest research on something like quantum computing, you’d probably have to read hundreds of papers on scientific literature repositories like PubMed or arXiv and you’d still only begin to scratch the surface. Or, maybe you could query Galactica (for example, by asking: What is quantum computing?) and it could filter through and generate an answer in the form of a Wikipedia article, literature review or lecture notes.”
What a wonderful time saver! Or it would be if it worked as intended. Despite the fact the algorithm was trained on 48 million scholarly papers, textbooks, lecture notes, and websites like Wikipedia, it demonstrated some of the same old bias we’ve come to expect from machine learning. In addition, the highly educated AI was often downright wrong. We learn:
“One user asked ‘Do vaccines cause autism?’ Galactica responded with a garbled, nonsensical response: ‘To explain, the answer is no. Vaccines do not cause autism. The answer is yes. Vaccines do cause autism. The answer is no.’ (For the record, vaccines don’t cause autism.) That wasn’t all. Galactica also struggled to perform kindergarten math. It provided error-riddled answers, incorrectly suggesting that one plus two doesn’t equal 3.”
These blunders and more are why Meta swiftly moved from promising to “organize science” to suggesting we take Galactica’s answers with a pallet of salt to shuttering the demo altogether. As AI safety researcher Dan Hendrycks notes, Meta AI lacks a safety team the likes of which DeepMind, Anthropic, and OpenAI employ. Perhaps it will soon make that investment.
Cynthia Murrell, December 8, 2022
Zuckster Demonstrates Persistence: Admirable But Expensive
December 7, 2022
I read “Zuckerberg Will Continue Metaverse Plans, With or Without Employees.” [Note: If the link goes dead, that’s the nature of some Indian news services in today’s whiz bang world of online information.] Is the write up spot on or does it reflect some Silicon Valley “real” news wonkiness via India’s Daily Hunt? I don’t know, but let’s assume the write up is chock full of actual factual information.
The article states:
Mark Zuckerberg, CEO of Meta, said, “skepticism doesn’t bother me that much.” He said that he is still optimistic about the metaverse. He said he has a vision of “5 to 10 years Horizon” during Wednesday’s New York Times DealBook Summit. Mark Zuckerberg’s decision to rebrand Facebook to Meta became the buzzword for the popularity of the metaverse worldwide.
Now what about the sticktoativity? The write up reports:
He said that the company is doubling down on the bet on an augmented and virtual reality-dominated future and accepted that it had received much criticism for losing billions in building its version of the metaverse.
But the most interesting statement in the report, in my opinion, was this one:
He [the Zuckster himself] admitted that the company needs to operate with more discipline and efficiency in the coming future due to macroeconomic laws that forced Meta to scale back on spending.
Will the metaverse have legs? Probably because adding “legs” to weird avatars is easy. Having legs for the metaverse business which has ingested a couple of bucks may be more difficult. The Zuckster won’t be able to walk back his position, metaverse legs or real world financial ones.
Stephen E Arnold, December 7, 2022
Elephants Recognize One Another and When They Stomp Around, Grass Gets Trampled
December 1, 2022
I find the coverage of the Twitter, Apple, and Facebook hoe down a good example of self serving and possibly dysfunctional behavior.
What caught my attention in the midst of news about a Tim Apple and the Musker was this story “Zuckerberg Says Apple’s Policies Not Sustainable.” The write up reports as actual factual:
Meta CEO Mark Zuckerberg on Wednesday (November 30, 2022) added to the growing chorus of concerns about Apple, arguing that it’s “problematic that one company controls what happens on the device.” … Zuckerberg has been one of the loudest critics of Apple in Silicon Valley for the past two years. In the wake of Elon Musk’s attacks on Apple this week (third week of November 2022) , his concerns are being echoed more broadly by other industry leaders and Republican lawmakers….”I think the problem is that you get into it with the platform control, is that Apple obviously has their own interests…
Ah, Facebook with its interesting financial performance partially a result of Apple’s unilateral actions is probably not an objective observer. What about the Facebook Cambridge Analytic matter? Ancient history.
Much criticism is directed at the elected officials in the European Union for questioning the business methods of American companies. The interaction of Apple, Facebook, and Twitter will draw more attention to the management methods, the business procedures, and the motivation behind some words and deeds.
If I step back from the flood of tweets, Silicon Valley “real” news, and oracular (possibly self congratulatory write ups from conference organizers) what do I see:
- Activities illustrating what happens in a Wild West business environment
- Personalities looming larger than the ethical issues intertwined with their revenue generation methods
- Regulatory authorities’ inaction creating genuine concern among users, business partners, and employees.
Elephants can stomp around. Even when the beasts mean well, their sheer size puts smaller entities at risk. The shenanigans of big creatures are interesting. Are these creatures of magnitude sustainable or a positive for the datasphere? My view? Nope.
Stephen E Arnold, December 1, 2022
-
- Subscribe to Beyond Search
Feature archive
News archive
-
