Google PR: An Explainer about Smart Software
January 19, 2023
One of Google’s big wizards packs a brain with the impact of MK 7 16? 50 caliber gun. Boom. Boom. Boom.
Google does “novel” cats. What does Chess.com’s Mittens have to say about these felines? Perhaps, Mittens makes humans move. Google makes “novel” cats sort of move. © Google, 2023.
Jeff Dean has trained his intellectual weapons on a certain viral star in the smart software universe. “Google Research, 2022 & Beyond: Language, Vision and Generative Models.” The main point of the essay / blog post / PR salvo is that Google has made transformational advances. Great things are coming from the Google.
The explanation of the hows of the great things consume about 7,000 words. For Google, that’s the equivalent of a digital War and Peace with a preface written by Henry James.
Here’s a passage which I circled in three different Googley colors:
We are working towards being able to create a single model that can understand many different modalities fluidly — understanding what each modality represents in context — and then actually generate different modes in that context. We’re excited by progress towards this goal! For example, we introduced a unified language model that can perform vision, language, question answering and object detection tasks in over 100 languages with state-of-the-art results across various benchmarks. In future applications, people can engage more senses to get computers to do what they want — e.g., “Describe this image in Swahili.” We’ve shown that on-device multi-modal models can make interacting with Google Assistant more natural. And we’ve demonstrated models that can, in various combinations, generate images, video, and audio controlled by natural language, images, and audio. More exciting things to come in this space!
Notice the phrase “progress towards this goal.” Notice the example “Describe this image in Swahili.” Notice the exclamation mark. Google is excited.
The write up includes Google’s jargonized charts and graphs; for example, “Preferred Metric Delta” and “SuperGLUE Score.” There is a graphic explaining multi-axis attention mechanism. And more.
Enough “catty” meta-commentary.
Here are several observations:
- Artificial intelligence is a fruit basket of methods, math, and malarkey. The fact that Google wants to pursue AI responsibly sounds good. What’s “responsible” mean? What’s artificial intelligence? These are difficult questions, and ones that are not addressed in the quasi-academic blog essay. Google has to sell advertising to keep the lights on and the plumbing in tip top shape… mostly. Seven thousand words is public relations, content marketing, and a response to the wild and crazy hyperbole about OpenAI changing the world. Okay, maybe after the lawyers, the regulators, the content copyright holders have figured out what is going on inside the allegedly open black boxes.
- If the reports from Davos are semi-accurate, Microsoft’s tie up with OpenAI and the idea of putting ChatGPT in Word makes me wonder if Microsoft Bob and Microsoft Clippy will return, allegedly smarter than before. Microsoft is riding a marketing wave and hoping to make money.
- Google is burdened with the albatross of Dr. Timnit Gebru and others who were transformed into former Googlers. What about Dr. Gebru’s legitimate concerns about baked in bias. When one sucks in content, the system does not know that content objects are more or less “better,” “right,” or distorted due to a spidering time out due to latency. The fact remains that Google terminated people who attempted to point out some foundational flaws in what the Google was doing.
Net net: The write up does not talk about Forward Forward methods. The write up does not talk about the likelihood that regulators in the European Union will be interested in what and how Google moves forward. Google is in the regulatory spot light. Will those regulators believe that Google can change its spots like the “novel” cats in the illustration? ChatGPT is something to get venture funders, entrepreneurs, and Davos executives to think positive thoughts. That does not mean the system will deliver. What about Mr. Brin’s self driving car prediction or the clever idea of solving death? Google may have to emulate in part Tesla, a company which allegedly faked the hands-off, full self-driving demo of its smart software. Seven thousand words means one thing to me:
‘The Google doth protest too much, methinks.’ Hamlet, Act 3, Scene 2. (I think Shakespeare put Google in a foul paper and some busybody inserted the name Gertrude.)
Boom, boom, boom.
Stephen E Arnold, January 19, 2023
Google, Take Two Aspirin and One AlkaSeltzer: It Is Buzz Time for ChatGPT
January 17, 2023
What do you do when the “trust” outfit Thomson Reuters runs a story with this headline? “Davos 2023: CEOs Buzz about ChatGPT-Style AI at World Economic Forum.” If you are like me, you think, “Meh.”
But what if you are a Google / DeepMind wizard?
Now consider this headline: “Google’s Muse Model Could Be the Next Big Thing for Generative AI.” If you are like me, you think, “Sillycon Valley PR.”
But what if you are an OpenAI or Microsoft brainiac?
In terms of reach, I think the Reuters’ story will be diffused to a broader business audience. The subject is something perceived as magnetic. Any carpetlander can get an associate to demonstrate ChatGPT outputting a search result via You.com or some other knowledge product from the numerous demos available with a mouse click.
But to see the Google Muse story, one has to follow a small number of Sillycon Valley outlets. And what if the carpetlander wants to see a demonstration of the magical, super effective Muse? Yeah, use your imagination.
Perhaps Google and its ineffable search team may want to crunch on another couple of aspirin and get some of that chewable antacid stuff. It is going to be a long PR day at Davos.
One doesn’t have to be a business maven to understand that ChatGPT is a nice subject when the options at Davos are war, plummeting demand for some big buck commodities, Germany’s burning lignite, China’s Covid and Taiwan fixation, and similar economically interesting topics.
What will CEO and Davos attendees take away from the ChatGPT buzz? My experience suggests some sort of action, even it is nothing more than investigating whether the technology can deal with pesky customer support inquiries.
And where is Google amidst this buzz? Google has the forward forward, next big thing. Google has academic papers which point out the weaknesses of non Google methods. Google has Muse or at least a news release story about Muse.
Will OpenAI and ChatGPT have legs? Who knows. Good bad or indifferent, ChatGPT has buzz, lots of it. I know because the “trust” outfit says ChatGPT will “transform” the security minded Microsoft. Who knew?
Thus, at this moment in time, Google may become a good customer for over-the-counter headache remedies and AlkaSeltzer. Remember that jingle’s lyrics?
Plop plop, fizz fizz
Oh, what a relief it is.
Maybe ChatGPT will just fade away like hangover or the tummy ache from eating the whole thing? Is it my imagination or is Microsoft chowing down on croissants whilst explaining what ChatGPT will do for its enterprise customers?
I will consult my “muse.” Oh, sorry, not available.
Stephen E Arnold, January 17, 2023
Smart Software KN Handle This Query for Kia
January 16, 2023
I read “People Can’t Read New Kia Logo, Resulting in 30,000 Monthly Searches for “KN Car“. The issue seems to be a new logo which when viewed by me seems to read the letter K and the letter N. Since I am a dinobaby, I assumed that I was at fault. But, no. The write up states:
All told, just 56% of the 1,062 survey participants nailed it, while 44% could not correctly identify the letters. Furthermore, 26% of respondents guessed it says “KN”—which results in roughly 30,000 online searches for “KN car” a month, according to Rerev.
I think this means that even the sharp eyed devils (my classification phrase for those in the GenX, GenY, and Millennial cohorts) cannot figure out the logo either.
I conjured up some of the marketing speak used to sell this new design to the Kia deciders:
- “Daddy, I know you hired me, and I like my new logo. You must make it happen. Okay, daddy.” — A person hired via nepotism
- “The dynamic lines make a bold statement about the thrust of the entire Kia line.” — From a bright eyed college graduate with a degree in business who is walking through the design’s advantages
- Okay. Modern?” — The statement by the Song Ho-sung after listening to everyone in the logo meeting.
To me, just change the name Kia to KM. BYD Auto may be a bigger problem than a KN logo.
Stephen E Arnold, January 16, 2023
Loving Tablets and Chromebooks: Sure, Like Going to the Dentist
December 29, 2022
Might smartphones make some devices irrelevant? We learn from The Register that “Tablet, Chromebook Shipments Come Crashing Down.” The article examines IDC’s report of third-quarter shipments. It states a mere 38.6 million tablets were shipped between July 01 and September 30, a decline of almost 9% since the previous year. Only Huawei grew its sales as demand escalated in China and Russia, where sanctions barred the way for Western tech. Writer Paul Kunert reports:
“Apple saw sales decline 1.1 percent to 14.5 million, according to IDC estimates. Samsung was down 4 percent to 7.1 million, Amazon fell 8.1 percent to 4.3 million, Lenovo shipments dropped 36.6 percent to 2.7 million, and Huawei grew 2 percent to 2.4 million. In its results filed late last week, Apple said iPad sales to end users were up 21 percent to $8.3 billion in Q4 of its fiscal ’22 ended 30 September despite supply constraints. IDC tracks sales into the channel, hence the difference in the figures. Chromebook shipments fell at a far faster rate, down 34.4 percent year-on-year to 4.3 million devices. This was the fifth straight decline for this sector of the PC industry. The downward trajectory began in the US, which accounted for 70 percent of global shipments. … IDC placed Acer as market leader with shipments of 1 million, albeit down 23.8 percent on a year ago. Dell shrank 19.9 percent to 900,000 units, HP was down 26.8 percent to 800,000, Lenovo plunged 54.8 percent to 700,000, and Samsung was down 37 percent to 300,000.”
Researchers point out Chromebook sales spiked during the pandemic as students connected from home, so its decline is simply a return to normal levels. As for the rest, a tough economy was likely at play. Apparently one can endure a slightly smaller small screen when fuel and groceries are difficult to afford.
The Arnold IT team has a different set of conclusions:
- Tablets and Chromebooks are like wearing clothing two sizes to small. Think discomfort.
- The promoters of tablets and Chromebooks are likely to use laptops to do “real” work.
- Tablets and Chromebooks make routine tasks difficult; for example, keeping an Internet connection in Buenos Aires during the World Cup Parade and finding a dongle in Hermanus.
Money and power allow some outfits to sell unusual stuff. Why not advertise these products on cable at 3 am?
Cynthia Murrell, December 29, 2022
Cyber Security: Is It Time for a Brazen Bull?
December 28, 2022
The cyber security industry has weathered Covid, mergers, acquisitions, system failures, and — excuse the lousy pun — solar winds. The flow of exploits with increasingly poetic names continues; for example, Azov, Zerobot, Killnet, etc. However, the cyber defense systems suffer from what one might call a slight misalignment. Bad actors find ways to compromise [a] humans to get user names and passwords, [b] exploit what is now the industry standard for excellence (MVP or minimal viable product, good enough engineering, and close-enough-for-horseshows technology), any gizmo or process connected to something connected to a public-facing network. The list of “bad” actors is a lengthy one. It includes bird-owning individuals in the UK, assorted government agencies hostile to the US, students in computer science class or hanging out in a coffee shop, and double agents with computing know how.
To add to the pain of cyber security, there are organizations which do great marketing but less great systems. “What’s in a PR Statement: LastPass Breach Explained” discusses a serious problem which underscores a number of issues.
LastPass is a product with a past reaching backwards more than a decade. The software made it easier for a user to keep track of what user name and password was whipped up to log into an online service or software. Over the years, PC Magazine found the password manager excellent. (Software can be excellent? Who knew?) Wikipedia has a list of “issues” the security software faced over the years. You can find that information here. More amusing is security expert Steve Gibson’s positive review of LastPass. Should you have the time, you can read about that expert’s conclusions in 2010 here.
But what does the PR statement article say? Here are a couple of snippets from the cited December 26, 2022, essay:
Snippet 1: Right before the holiday season, LastPass published an update on their breach. As people have speculated, this timing was likely not coincidental but rather intentional to keep the news coverage low. …Their statement is also full of omissions, half-truths and outright lies.
Harsh.
Snippet 2: Again, it seems that LastPass attempts to minimize the risk of litigation (hence alerting businesses) while also trying to prevent a public outcry (so not notifying the general public). Priorities…
My take on LastPass is that the company is doing what other cyber security firms do: Manage information about problems.
Let’s talk about cyber security on a larger stage. How does a global scale sound?
First, security is defined by [a] what bad actors have been discovered to do and [b] marketing. A breach occurs. A fix — ideally one enabled by artificial intelligence and chock full of predictive analytics — is created and marketed. Does the fix work? How about those Exchange Server exploits or those 24×7 phishing attacks? The point for me is that cyber security seems to be reactive; that is, dictated by what bad actors do.
Second, the “fix” is verified by whom and what? In the US there are Federal cyber groups. There are state cyber groups. There are cyber associations. There are specialty labs in fun places like Quantico. For a LastPass incident, which cowpoke moves the cow along? The point: Bureaucracy, friction, artificial barriers, time, expertise, money, and more.
Third, technical layoffs and time mean that cyber crime may be an attractive business opportunity for some.
Considering these three points, I want to hazard several observations:
- Cyber security may be an oxymoron
- Bad actors have the advantages granted by good enough software and systems, tools, talent, and time
- Users and customers who purchase security may be faced with a continual flow of surprises
What’s the fix? May I suggest that we consider bringing back the Bull of Phalaris aka the brazen bull.
The “bull” is fabricated of a suitable metal; for example, bronze. The inside of the bull is hollow. A trapdoor allows access to the interior space. When the trapdoor is closed, there is an opening from the interior to the bull’s nose. The malefactor — let’s say a venture firm’s managing director who is rolling up cyber security companies with flawed software — is placed inside the bull. A fire is built beneath the bull and the shouts and possible other noises are emitted from the opening in the bull’s head.
The use of the brazen bull for software developers pumping out “good enough” cyber security solutions can be an option as well. Once law enforcement snags the head of a notorious hacking gang, the bull will be pressed into duty. Keep in mind that Microsoft blamed 1,000 cyber warriors working in a country hostile to the US for the SolarWinds’ misstep. This would necessitate more bulls which would provide meaningful work to some.
I would advocate that marketer types who sell cyber security systems which don’t work be included in the list of individuals who can experience the thrill of the brazen bull.
My thought is that the use of the brazen bull with clips released as short videos would capture some attention.
What’s is going on now is not getting through? More robust measures are necessary. No bull.
Stephen E Arnold, December 28, 2022
Want Clicks? Put War Videos on TikTok
December 20, 2022
Here is another story about the importance of click-throughs to social media companies, repercussions be damned. BBC News reports, “Russian Mercenary Videos ‘Top 1 Bn Views’ on TikTok.” The mercenary band in these videos, known as the Wagner Group, is helping Russia fight its war against Ukraine. Writer Alexandra Fouché cites a recent report from NewsGuard as she reveals:
“NewsGuard said it had identified 160 videos on the short-video platform that ‘allude to, show, or glorify acts of violence’ by the mercenary group, founded by Yevgeny Prigozhin, a close ally of President Vladimir Putin. Fourteen of those videos showed full or partial footage of the apparent killing of former Russian mercenary Yevgeny Nuzhin which saw high engagement within days of being uploaded last month, it said.”
That brutal murder, which was performed with a sledgehammer, was viewed over 900,000 times on TikTok before ByteDance took it down. Nuzhin was apparently killed because he switched sides and denounced the Wagner Group. Sadly but surely, there are many viewers who would seek out such footage; why blame TikTok for its spread? The article continues:
“NewsGuard found that TikTok’s algorithm appeared to push users towards violent Wagner Group content. When an analyst searched for the term ‘Wagner’, TikTok’s search bar suggested searches for ‘Wagner execution’ and ‘Wagner sledgehammer’. The same search in Russian resulted in the suggestions ‘Wagner PMC’, ‘Wagner sledgehammer’ and ‘Wagner orchestra’. Wagner refers to its fighters as ‘musicians’. NewsGuard also found that videos could be found on TikTok showing another Wagner murder involving an army deserter in Syria in 2017 and that they had reached millions of users.
The online analysis group said it had also identified other music videos on the platform that advocated violence against Ukrainians, including calls to kill Ukrainians claiming they were ‘Nazis’.
Funny, when I searched Google for “Wagner,” the first three results my filter bubble turned up were composer Richard Wagner’s Wikipedia page, Wagner paint sprayer’s home page, and Staten Island’s Wagner College. Some actual news articles about the Wagner Group followed, but nary a violence glorification video in sight. TikTok certainly knows how to generate clicks. But what about China’s “reeducation” camps? The Chinese company is not circulating videos of those, is it? It seems the platform can be somewhat selective, after all.
Cynthia Murrell, December 20, 2022
Over the Holidays Learn Algospeak
December 19, 2022
Internet content has evolved its own set of coded words and emojis called algospeak. Though often discernable from context, the meanings behind these terms and symbols can easily escape the uninitiated. Lifehacker supplies a list of such terms in, “All the Social Media ‘Algospeak’ You Don’t Understand.” But wait, you might ask, what is wrong with clarity and accurate wording? Writer Sarah Showfety explains:
“If you’ve ever created content for internet consumption, you know the number one rule: Don’t upset the algorithm. Modern-day social media algorithms are like the Wizard of Oz—cloaked, all-powerful puppeteers who can seemingly perform miracles for the right creators, instantly propelling their content in front of millions of eyeballs. But they are as fickle as they are promising, often trapping content in a dungeon of 53 views for no discernible reason. While the inner machinations of algorithms are largely unknown, being blackballed by one can suppress your content and seal your doom—and one of the quickest ways to do that is to use language that could be flagged as a violation of the platform’s content guidelines or terms of service. So internet content creators have developed a growing glossary of terms designed to circumvent automated brand safety filters. This evolving lexicon of euphemisms, abbreviations, deliberate misspellings, symbol insertions and emojis known as ‘algospeak’ is used to disguise sensitive and potentially problematic words having to do with polarizing political topics, controversial global events, cultural taboos, death, drugs, and just plain sex.”
Ah yes, the almighty algorithm. Readers may want to bookmark the post in case of future confusion. A few of the PG-rated euphemisms include “bink in lio” for “link in bio,” “swimmers” for vaccinated people, and the sunflower emoji to symbolize Ukraine. Showfety points out one entry is particularly unfortunate, at least for this charismatic kid who suddenly found himself famous for his love of maize: “Corn” is algospeak for “porn.” Really? We don’t make these choices, we just try to keep you informed, dear reader.
Cynthia Murrell, December 19, 2022
Google to Microsoft: We Are Trying to Be Helpful
December 16, 2022
Ah, those fun loving alleged monopolies are in the news again. Microsoft — famous in some circles for its interesting approach to security issues — allegedly has an Internet Explorer security problem. Wait! I thought the whole wide world was using Microsoft Edge, the new and improved solution to Web access.
According to “CVE-2022-41128: Type Confusion in Internet Explorer’s JScript9 Engine,” Internet Explorer after decades of continuous improvement and its replacement has a security vulnerability. Are you still using Internet Explorer? The answer may be, “Sure you are.”
With Internet Explorer following Bob down the trail of Microsoft’s most impressive software, the Redmond crowd the Microsoft Office application uses bits and pieces of Internet Explorer. Thrilling, right?
Google explains the Microsoft issue this way:
The JIT compiler generates code that will perform a type check on the variable
q
at the entry of theboom
function. The JIT compiler wrongly assumes the type will not change throughout the rest of the function. This assumption is broken whenq
is changed fromd
(anInt32Array
) toe
(anObject
). When executingq[0] = 0x42424242
, the compiled code still thinks it is dealing with the previousInt32Array
and uses the corresponding offsets. In reality, it is writing to wherevere.e
points to in the case of a 32-bit process ore.d
in the case of a 64-bit process. Based on the patch, the bug seems to lie within a flawed check inGlobOpt::OptArraySrc
, one of the optimization phases.GlobOpt::OptArraySrc
callsShouldExpectConventionalArrayIndexValue
and based on its return value will (in some cases wrongly) skip some code.
Got that.
The main idea is that Google is calling attention to the future great online game company’s approach to software engineering. In a word or two, “Poor to poorer.”
My view of the helpful announcement is that Microsoft Certified Professionals will have to explain this problem. Google’s sales team will happily point out this and other flaws in the Microsoft approach to enterprise software.
If you can’t trust a Web browser or remove flawed code from a widely used app, what’s the fix?
Ready for the answer: “Helpful cyber security revelations that make the online ad giant look like a friendly, fluffy Googzilla. Being helpful is the optimal way to conduct business.
Stephen E Arnold, December 16, 2022
Rainbow Narcotics? Just a Coincidence? Nope, Marketing Plain and Simple
December 9, 2022
Does anyone remember how the tobacco companies had ads and mascots that appealed to kids but claimed more than once their target demographic wasn’t children? It was a bald-faced lie as big as the former claim that smoking does not negatively affect health. The Daily Caller has a whopper of a story about fentanyl: “Drug Cartel Operative Claims Rainbow Fentanyl Was Not Created To ‘Make Kids Addicts.’”
A Mexican drug cartel operative told Insider that rainbow-colored fentanyl is not meant to make kids addicts. The fentanyl pills have the same colors and shapes as popular candies such as Smarties, Sweet Tarts, and more. The cartel operative said the bright colors are meant to warn adults the pills contain fentanyl:
“‘We know that some of the dealers in the US started mixing cocaine with ‘fenta’ without letting their buyers know, and that is very dangerous,’ the operative told Insider. The colorful drug form was created ‘to make it look different than coke or white heroin,’ a Sinaloa cartel drug cook explained, according to Insider. ‘Also we mix some of the heroin with fentanyl to make it more powerful, but we mark it, to let the buyer know that this one has ‘fenta,’’ the operative added. ‘Whatever happens when it’s taken from our hands, it’s not our problem.’”
Ann Milgram, the Administrator of the Drug Enforcement Administration (DEA), pushes back against the cartels by stating the opposite. She claims that rainbow fentanyl is a deliberate attempt to target American kids.
The same cartel operative says they cook clean “fenta” and clearly label it with “el arco del iris” (rainbow).
Right. And Joe Camel was as friendly as Chuck E. Cheese, McGruff the Crime Dog, Smokey the Bear, and Ronald McDonald.
Whitney Grace, December 9, 2022
Can Clever Smart Software Identify Misinformation?
December 8, 2022
My view is, “Nope.” What will marketers say? My thought is, “Anything, anything at all.”
Navigate to “Physicists Create a Wormhole Using a Quantum Computer.” Read it. Now click on “The Death of Quanta Magazine” and read the essay about the Wormhole write up. Here’s the question: “Can you identify the misinformation in each essay?” The $64 dollar question is: “Can smart software flag and tag the misinformation?”
My hunch is that most humans, even the highly intelligent ones reading my article about these two essays, will have a difficult time identifying factoids, hypotheses, and baloney. Now is smart software from one of the allegedly open source outfits or a rapacious but user friendly commercial service able to handle this task?
Let’s look at one passage from the “The Death of Quanta Magazine”; to wit:
While the article correctly points out that one needs negative energy to make a wormhole traversable, and that negative energy does not exist, and that the experiment merely simulated a negative energy pulse, the video has no such qualms. It directly stated that the experiment created a negative energy shockwave and used it to transmit qubits through the wormhole. For me the worst part of the video was at 11:53, where they showed a graph with a bright point labeled “negative energy peak” on it. The problem is that this is not a plot of data, it’s just a drawing, with no connection to the experiment. Lay people will think they are seeing actual data, so this is straightforward disinformation.
Several observations:
- An article and a video. The combo suggests that presumably intelligent people writing about what is allegedly a scientific presentation are chasing the ethos of TikTok and YouTube. Interesting but didn’t Newton get along with a pen and paper?
- Fancy lingo. Yep, holograms, sci-fi sounding jargon like negative energy, and obligatory static graphs.
- Experts. Wow. Experts offered up without much context. Impressive indeed.
- Meta-commentary. I love it when articles comment on other articles. Great fun.
The problem is that smart software may struggle with the nuances in the two articles. Quanta will do an article about that soon I expect.
Content marketing, pseudo tech baloney, and clicks. Yeah.
Stephen E Arnold, December 8, 2022