New Wave Management or Is It Leaderment?
September 12, 2023
Note: This essay is the work of a real and still-alive dinobaby. No smart software involved, just a dumb humanoid.
Here’s one of my biases, and I am rather proud of it. I like to word “manager.” According to my linguistics professor Lev Soudek, the word “manage” used to mean trickery and deceit. When I was working at a blue chip consulting firm, the word meant using tactics to achieve a goal. I think of management as applied trickery. The people whom one pays will go along with the program, but not 24×7. In a company which expects 60 hours of work a week the minimum for survival of a Spanish inquisition inspired personnel approach, mental effort had to be expended.
I read “I’m a Senior Leader at Amazon and Have Seen Many Bad Managers. Here Are 3 Reasons Why There Are So Few Great Ones.” The intense, clear-eyed young person explains that he has worked at some outfits which are not among my list of the Top 10 high-technology outfits. His résumé includes eBay (a digital yard sale), a game retailer, and the somewhat capricious Amazon (are we a retail outfit, are we a cloud outfit, are we a government services company, are we a data broker, are we a streaming company, etc.).
A modern practitioner of leaderment is having trouble getting the employees to fall in, throw their shoulders back, and mark in step to the cadence of Am-a-zon, Am-a-zon like a squad of French Foreign Legion troops on Bastille Day. Thanks, MidJourney. The illustration did not warrant a red alert, but it is also disappointing.
I assume that these credentials are sufficient to qualify for a management guru. Here are the three reasons managers are less than outstanding.
First, managers just sort of happen. Few people decide to be a manager. Ah, serendipity or just luck.
Second, managers don’t lead. (Huh, the word is “management”, not “leaderment.”)
Third, pressure for results means some managers are “sacrificing employee growth.” (I am not sure what this statement means. If one does not achieve results, then that individual and maybe his direct reports, the staff he leaderments, and his boss will be given an opportunity to find their future elsewhere. Translation for the GenZ reader: You are fired.
Let’s step back and think about these insights. My initial reaction is that a significant re-languaging has taken place in the write up. A good manager does not have to be a leader. In fact, when I was a guest lecturer at the Kansai Institute of Technology, I met a number of respected Japanese managers. I suppose some were leaders, but a number made it clear that results were number one or ichiban.
In my work career, confusing to manage with to lead would create some confusion. I recall when I was working in the US Congress with a retired admiral who was elected to represent an upscale LA district, the way life worked was simple: The retired admiral issued orders. Lesser entities like myself figured out how to execute, tapped appropriate resources, and got the job done. There was not much leadership required of me. I organized; I paid people money; and I hassled everyone until the retired admiral grunted in a happy way. There was no leaderment for me. The retired admiral said, “I want this in two days.” There was not much time for leaderment.
I listened to a podcast called GeekWire. The September 2, 2023, program made it clear that the current big dog at Amazon wants people to work in the office. If not, these folks are going to go away. What makes this interesting is that the GeekWire pundits pointed out that the Big Dog had changed his story, guidelines, and procedures for this work from home and work from office approach multiple times.
Therefore, I am not sure if there is management or leaderment at the world’s largest digital mall. I do know that modern leaderment is not for me. The old-fashioned meaning of manage seems okay to me.
Stephen E Arnold, September 12, 2023
Google: An Ad Crisis Looms from the Cancer of Short Videos
September 7, 2023
The weird orange newspaper ran a story which I found important. To read the article, you will need to pony up cash; I suggest you consider doing that. I want to highlight a couple of key points in the news story and offer a couple of observations.
An online advertising expert looks out his hospital window and asks, “I wonder if the cancer in my liver will be cured before the cancer is removed from my employer’s corporate body?” The answer may be, “Liver cancer can be has a five year survival rate between 13 to 43 percent (give or take a few percentage points).” Will the patient get back to Foosball and off-site meetings? Is that computer capable of displaying TikTok videos? Thanks, Mother MJ. No annoying red appeal this banners today.
The article “Shorts Risks Cannibalising Core YouTube Business, Say Senior Staff” contains an interesting (although one must take with a dollop of mustard and some Dead Sea salt):
Recent YouTube strategy meetings have discussed the risk that long-form videos, which produce more revenue for the company, are “dying out” as a format, according to these people.
I am suspicious of quotes from “these people.” Nevertheless, let’s assume that the concern at the Google is real like news from “these people.”
The idea is that Google has been asleep at the switch as TikTok (the China linked short video service) became a go-to destination for people seeking information. Yep, some young people search TikTok for information, not just tips on self-harm and body dysmorphia. Google’s reaction was slow and predictable: Me too me too me too. Thus, Google rolled out “Shorts,” a TikTok clone and began pushing it to its YouTube faithful.
The party was rolling along until “these people” sat down and looked at viewing time for longer videos and the ad revenue from shorter videos. Another red alert siren began spinning up.
The orange newspaper story asserted:
In October last year, YouTube reported its first-ever quarterly decline in ad revenue since the company started giving its performance separately in 2020. In the following two quarters, the platform reported further falls compared with the same periods the previous year.
With a decline in longer videos, the Google cannot insert as many ads. If people watch shorter videos, Google has reduced ad opportunities. Although Google would love to pump ads into 30 second videos, viewers (users) might decide to feed their habit elsewhere. And where one may ask? How about TikTok or the would be cage fighter’s Meta service?
Several observations:
- Any decline in ad revenue is a force multiplier at the Google. The costs of running the outfit are difficult to control. Google has not been the best outfit in the world in creating new, non ad revenue streams in the last 25 years. That original pay-to-play inspiration has had legs, but with age, knees and hips wear out. Googzilla is not as spry as it used to be and its bright idea department has not found sustainable new revenue able to make up for a decline in traditional Google ad revenue… yet.
- The cost of video is tough to weasel out of Google’s financial statements. The murky “cloud” makes it easy to shift some costs to the enabler of the magical artificial intelligence push at the company. In reality, video is a black hole of costs. Storage, bandwidth, legal compliance, creator hassles, and overhead translate to more ads. Long videos are one place to put ads every few minutes. But when the videos are short like those cutting shapes dance lessons, the “short” is a killer proposition.
- YouTube is a big deal. Depending on whose silly traffic estimates one believes, YouTube is as big a fish in terms of eyeballs as Google.com search. Google search is under fire from numerous directions. Prabhakar Raghavan has not mounted much of a defense to the criticisms directed at Google search’s genuine inability to deliver relevant search results. Now the YouTube ad money flow is drying up like streams near Moab.
Net net: YouTube has become a golden goose. But short videos are a cancer and who can make fois gras out of a cancerous liver?
Stephen E Arnold, September 7, 2023
Search: The Moonshot for Alphabet Google YouTube Etc. May Be Off by Miles
September 6, 2023
Note: This essay is the work of a real and still-alive dinobaby. No smart software involved, just a dumb humanoid.
Google is now 25. Yep, a quarter century. If you want to read a revisionist history of the beloved firm, point your Chrome browser (yep, it is part of the alleged monopoly) at “Questions, Shrugs and What Comes Next: A Quarter Century of Change.” The cited article appears in the Google blog (does anyone remember Blogger or what about the Google blog search?). The idea is that Sundar Pichai logged into a Google workspace (yep, that’s the software system intended to deal Microsoft a mortal blow).
I just wanted to hire a normal clown. It was not possible. The clown search became a monster. Let’s have fun! Thanks, MidJourney, you old gradient descent entity.
What does Mr. Pichai write, allegedly without the assistance of other Googlers, advisors, and legal eagles?
One of this statements is:
Search is still at the core of our mission, and it’s still our biggest moonshot with so much more to do.
Okay, I want to stop there. I wanted to find a service in Louisville, Kentucky, that sends clowns to birthday parties. Pretty simple, right. I entered the query “Louisville Kentucky clowns birthday parties.” I expected to see a list of people or companies in the clown rental business. Wrong? I received this output from the “biggest moonshot” outfit:
The top hit was to The Bash, a service which lists clowns. That link pointed me to Bunny Bear Entertainment and Salem Sisters 502. No phone number, just a link to get a free quote. Okay, that looks good. Click on the link and what happens? A form appears and someone will contact me. Yeah, I wanted to talk to a person.
The second hit was to Phillips (presenting itself as kiddyskingdom.com) at a toll free number. Same deal. A referral service. No, I could not talk to a human in Louisville.
The third hit was to About Faces. Yep, another SEO-ized reseller of clown services. No phone number for me to call to talk to a real live clown.
Several observations:
- Google search (the moonshot) is not useful. It creates work; it does not provide what I wanted.
- Google’s business is selling ads which are funding Google Cloud ambitions to break out of the one-trick-pony pejorative aimed at the company by the Softie Steve Ballmer a long time ago.
- The blog post is a marketing pitch for Google’s smart software.
Net net: Vintage Google operating without regard to regulatory scrutiny, allegations that the company is a monopoly, or managing people in a way that is what I hoped the clown company would provide to me: Laughs.
A “healthy disregard for the impossible.” Sure. I trust Google. I believe the Google. But delivering on point search results. I don’t need a clown for that. I have one.
Stephen E Arnold, September 6, 2023
Amazon Offers AI-Powered Review Consolidation for Busy Shoppers
September 6, 2023
I read the reviews for a product. I bought the product. Reality was — how shall I frame it — different from the word pictures. Trust those reviews. ? Hmmm. So far, Amazon’s generative AI focus has been on supplying services to developers on its AWS platform. Now, reports ABC News, “Amazon Is Rolling Out a Generative AI Feature that Summarizes Product Reviews.” Writer Haleluya Hadero tells us:
“The feature, which the company began testing earlier this year, is designed to help shoppers determine at a glance what other customers said about a product before they spend time reading through individual reviews. It will pick out common themes and summarize them in a short paragraph on the product detail page.”
A few mobile shoppers have early access to the algorithmic summaries while Amazon tweaks the tool with user feedback. Eventually, the company said, shoppers will be able to surface common themes in reviews. Sounds nifty, but there is one problem: Consolidating reviews that are fake, generated by paid shills, or just plain wrong does nothing to improve their accuracy. But Amazon is more eager to jump on the AI bandwagon than to perform quality control on its reviews system. We learn:
“The Seattle-based company has been looking for ways to integrate more artificial intelligence into its product offerings as the generative AI race heats up among tech companies. Amazon hasn’t released its own high-profile AI chatbot or imaging tool. Instead, it’s been focusing on services that will allow developers to build their own generative AI tools on its cloud infrastructure AWS. Earlier this year, Amazon CEO Andy Jassy said in his letter to shareholders that generative AI will be a ‘big deal’ for the company. He also said during an earnings call with investors last week that ‘every single one’ of Amazon’s businesses currently has multiple generative AI initiatives underway, including its devices unit, which works on products like the voice assistant Alexa.”
Perhaps one day Alexa will recite custom poetry or paint family portraits for us based on the eavesdropping she’s done over the years. Heartwarming. One day, sure.
Cynthia Murrell, September 19, 2023
Google: Trapped in Its Own Walled Garden with Lots of Science Club Alums
August 30, 2023
Note: This essay is the work of a real and still-alive dinobaby. No smart software involved, just a dumb humanoid.
I read “MapReduce, TensorFlow, Vertex: Google’s Bet to Avoid Repeating History in AI.” I found the idea that Google gets in its own way a retelling of how high school science club management produces interesting consequences.
A young technology wizard finds himself in a Hall of Mirrors at the carnival. He is not sure what is real or in which direction to go. The world of the House of Mirrors is disorienting. The young luminary wants to return to the walled garden where life is more comfortable. Thanks, MidJourney. Four tries and I get this tired illustration. Gradient descent time?
The write up asserts:
Google is in the middle of trying to avoid repeating history when releasing its industry-altering technology.
I disagree. The methods defining Google produce with remarkable consistency a lack of informed control. The idea is that organizations have a culture. That cultural evolves over time, but it remains anchored in its past. Thus, as the organization appears to move forward in time, that organization behaves in a predictable way; for example, Google has an approach to management which guarantees friction. Examples range from the staff protests to the lateral arabesque used to move Dr. Jeff Dean out of the way of the DeepMind contingent.
The write up takes a different view; for example:
Run by engineers, the [Google MapReduce] team essentially did not foresee the coming wave of open-source technology to power the modern Web and the companies that would come to commercialize it.
Google lacks the ability to perceive its opportunities. The company is fenced by its dependence on online advertising. Thus, innovations are tough for the Googlers to put into perspective. One reason is the high school science club ethos of the outfit; the other is that the outside world is as foreign to many Googlers as the world beyond the goldfish’s bowl filled with water. The view is distorted, surreal, and unfamiliar.
How can a company innovate and make a commercially viable product with this in its walled garden? It cannot. Advertising at Google is a me-too product for which Google prior to its IPO settled a dispute with Yahoo over the “inspiration” for pay-to-play search. The cost of this “inspiration” was about $1 billion.
In a quarter century, Google remains what one Microsoftie called “a one-trick pony.” Will the Google Cloud emerge as a true innovation? Nope. There are lots of clouds. Google is the Enterprise Rent-a-Car to the Hertz and Avis cloud rental firms. Google’s innovation track record is closer to a high school science club which has been able to win the state science club content year after year. Other innovators win the National Science Club Award (once called the Westinghouse Award). The context-free innovations are useful to others who have more agility and market instinct.
My view is that Google has become predictable, lurching from one technical paper to legal battle like a sine wave in a Physics 101 class; that is, a continuous wave with a smooth periodic function.
Don’t get me wrong. Google is an important company. What is often overlooked is the cultural wall that keeps the 100,000 smartest people in the world locked down in the garden. Innovation is constrained, and the excitement exists off the virtual campus. Why do so many Xooglers innovate and create interesting things once freed from the walled garden? Culture has strengths and weaknesses. Google’s muffing the bunny, as the article points out, is one defining characteristic of a company which longs for high school science club meetings and competitions with those like themselves.
Tony Bennett won’t be singing in the main cafeteria any longer, but the Googlers don’t care. He was an outsider, interesting but not in the science club. If the thought process doesn’t fit, you must quit.
Stephen E Arnold, August 30. 2023
MBAs Want to Win By Delivering Value. It Is Like an Abstraction, Right?
August 11, 2023
Note: This essay is the work of a real and still-alive dinobaby. No smart software involved, just a dumb humanoid.
Is it completely necessary to bring technology into every aspect of one’s business? Maybe, maybe not. But apparently some believe such company-wide “digital transformation” is essential for every organization these days. And, of course, there are consulting firms eager to help. One such outfit, Third Stage Consulting Group, has posted some advice in, “How to Measure Digital Transformation Results and Value Creation.” Value for whom? Third Stage, perhaps? Certainly, if one takes writer Eric Kimberling on his invitation to contact him for a customized strategy session.
Kimberling asserts that, when embarking on a digital transformation, many companies fail to consider how they will keep the project on time, on budget, and in scope while minimizing operational disruption. Even he admits some jump onto the digital-transformation bandwagon without defining what they hope to gain:
“The most significant and crucial measure of success often goes overlooked by many organizations: the long-term business value derived from their digital transformation. Instead of focusing solely on basic reasons and justifications for undergoing the transformation, organizations should delve deeper into understanding and optimizing the long-term business value it can bring. For example, in the current phase of digital transformation, ERP [Enterprise Resource Planning] software vendors are pushing migrations to new Cloud Solutions. While this may be a viable long-term strategy, it should not be the sole justification for the transformation. Organizations need to define and quantify the expected business value and create a benefits realization plan to achieve it. … Considering the significant investments of time, money, and effort involved, organizations should strive to emerge from the transformation with substantial improvements and benefits.”
So companies should consider carefully what, if anything, they stand to gain by going through this process. Maybe some will find the answer is “nothing” or “not much,” saving themselves a lot of hassle and expense. But if one decides it is worth the trouble, rest assured many consultants are eager to guide you through. For a modest fee, of course.
Cynthia Murrell, August 11, 2023
IBM and Smart Software: Try and Try Again, Dear Watson with an X
August 7, 2023
Note: This essay is the work of a real and still-alive dinobaby. No smart software involved, just a dumb humanoid.
With high hopes, IBM is acquiring FinOps firm Apptio for $4.6 billion. As Horses for Sources puts it, “IBM’s Acquisition of Apptio Can Shine if IBM Software and IBM Consulting Work Together to Deliver Cost-Managed Innovation at Speed.” But that is a big “if”. The odds seem long from the standpoint of RedHat users unimpressed with both IBM’s approach and internal cooperation at the company.
The young, sincere child presages her future in a giant technology company, “Yes, I will try to stack blocks to make the big building you told me to create with the blocks I got from my friend, Apt Ti Oh.” MidJourney, you did let me down with your previous “frustrated kid” images. Sultry teens were not what I was after.
IBM intends to mix Apptio with several other acquisitions that have gone into the new Watsonx platform, like Turbonomic, Instana, and MyInvenio, to create a cohesive IT-management platform. Linking spending data with operational data should boost efficiency, save money, and facilitate effective planning. This vision, however, is met with some skepticism. Writers Tom Reuner and Phil Fersht tell us:
“Apptio never progressed beyond providing insights, while IBM needs to demonstrate the proof points for integrating its disparate capabilities as well as progress from insight to action and, ultimately, automation. IBM Software must work with IBM Consulting transformation more effectively. … In essence, if successful, the ability to act on – and ultimately automate – all those insights is pretty much the operational Holy Grail. Just for transparency, getting expansive spend management and FinOps capabilities in itself will be a solid asset for IBM. However, any new and bolder proposition aiming at the bigger transformation price must move beyond technology and include stakeholders and change management. The ambition could be a broader business assurance where spend data, operational insights, and governance get tied to business objectives. In our view, this provides a significant alignment opportunity with IBM Consulting as it seeks to differentiate itself from the likes of Accenture Operations and Genpact. Having a deep services alignment with Watsonx and Apptio will bridge together the ability to manage the cost and value of both cloud transformation and AI investments – provided it gets it right with its global talent base of technical and process domain specialists.”
So the objective is a platform that brings companies’ disparate parts together into a cohesive and efficient whole. But this process must involve humans as well as data. If IBM can figure out how to do so within its own company, perhaps it stands a chance of reaching the goal.
Cynthia Murrell, August 6, 2023
When Wizards Flail: The Mysteries of Smart Software
July 18, 2023
![Vea4_thumb_thumb_thumb_thumb_thumb_t[1]](http://arnoldit.com/wordpress/wp-content/uploads/2023/07/Vea4_thumb_thumb_thumb_thumb_thumb_t1-30.gif "Vea4_thumb_thumb_thumb_thumb_thumb_t[1]")
Note: This essay is the work of a real and still-alive dinobaby. No smart software involved, just a dumb humanoid.
How about that smart software stuff? VCs are salivating. Whiz kids are emulating Sam AI-man. Users are hoping there is a job opening for a Wal-Mart greeter. But there is a hitch in the git along; specifically, some bright experts are not able to understand what smart software does to generate output. The cloud of unknowing is thick and has settled over the Land of Obfuscation.
“Even the Scientists Who Build AI Can’t Tell You How It Works” has a particularly interesting kicker:
“We built it, we trained it, but we don’t know what it’s doing.”
A group of artificial intelligence engineers struggling with the question, “What the heck is the system doing?” A click of the slide rule for MidJourney for this dramatic depiction of AI wizards at work.
The write up (which is an essay-interview confection) includes some thought-provoking comments. Here are three; you can visit the cited article for more scintillating insights:
Item 1: “… with reinforcement learning, you say, “All right, make this entire response more likely because the user liked it, and make this entire response less likely because the user didn’t like it.”
Item 2: “… The other big unknown that’s connected to this is we don’t know how to steer these things or control them in any reliable way. We can kind of nudge them
Item 3: “We don’t have the concepts that map onto these neurons to really be able to say anything interesting about how they behave.”
Item 4: “… we can sort of take some clippers and clip it into that shape. But that doesn’t mean we understand anything about the biology of that tree.”
Item 5: “… because there’s so much we don’t know about these systems, I imagine the spectrum of positive and negative possibilities is pretty wide.”
For more of this type of “explanation,” please, consult the source document cited above.
Several observations:
- I like the nudge and watch approach. Humanoids learning about what their code does may be useful.
- The nudging is subjective (human skill) and the reference to growing a tree and not knowing how that works exactly. Just do the bonsai thing. Interesting but is it efficient? Will it work? Sure or at least as Silicon Valley thinking permits
- The wide spectrum of good and bad. My reaction is to ask the striking writers and actors what their views of the bad side of the deal is. What if the writers get frisky and start throwing spit balls or (heaven forbid) old IBM Selectric type balls. Scary.
Net net: Perhaps Google knows best? Tensors, big computers, need for money, and control of advertising — I think I know why Google tries so hard to frame the AI discussion. A useful exercise is to compare what Google’s winner in the smart software power struggle has to say about Google’s vision. You can find that PR emission at this link. Be aware that the interviewer’s questions are almost as long at the interview subject’s answers. Does either suggest downsides comparable to the five items cited in this blog post?
Stephen E Arnold, July 18, 2023
Microsoft Causing Problems? Heck, No
July 14, 2023
![Vea4_thumb_thumb_thumb_thumb_thumb_t[1]](http://arnoldit.com/wordpress/wp-content/uploads/2023/07/Vea4_thumb_thumb_thumb_thumb_thumb_t1-27.gif "Vea4_thumb_thumb_thumb_thumb_thumb_t[1]")
Note: This essay is the work of a real and still-alive dinobaby. No smart software involved, just a dumb humanoid.
I cruised through the headlines my smart news system prepared for me. I noted two articles on different subjects. The two write ups were linked with a common point of reference: Microsoft Corp., home of the Softies and the throbbing heart of a significant portion of the technology governments in North America and Western Europe find essential.
“What’s the big deal?” asks Mr. Microsoft. “You have Windows. You have Azure. Software has bugs. Get used to it. You can switch to Linux anytime.” Thin interesting scene is the fruit of MidJourney’s tree of creativity.
The first article appeared in TechRadar. an online real news outfit. The title was compelling; specifically, “Windows 11 Update Is Reportedly Slowing Down PCs and Breaking Internet Connections.” The write up reports:
KB5028185, the ‘Moment 3’ update, is proving seriously problematic for some users … The main bones of contention with patch KB5028185 for Windows 11 22H2 are instances of performance slowdown – with severe cases going by some reports – and problems with flaky internet connections.
The second story appeared on cable “real” news. I tracked down the item titled “US and Microsoft Sound Alarm about China-Based Cybersecurity Threat.” The main idea seems to be:
The U.S. and Microsoft say China-based hackers, focused on espionage, have breached email accounts of about two dozen organizations, including U.S. government agencies.
Interesting. Microsoft seems to face two challenges: Desktop engineering and cloud engineering. The common factor is obviously engineering.
I am delighted that Bing is improving with smart software. I am fascinated by Microsoft’s effort to “win” in online games. However, isn’t it time for something with clout to point out that Microsoft may need to enhance its products’ stability, security, and reliability.
Due to many organizations’ and individuals’ dependence on Microsoft, the company seems to have a knack for creating a range of issues. Will someone step up and direct the engineering in a way that does not increase vulnerability and cause fiduciary loss for its customers?
Anyone? Crickets I fear. Bad actors find Microsoft’s approach more satisfying than a stream of TikTok moments.
Stephen E Arnold, July 14, 2023
Open AI and Its Alignment Pipeline
July 12, 2023
![Vea4_thumb_thumb_thumb_thumb_thumb_t[1]](http://arnoldit.com/wordpress/wp-content/uploads/2023/07/Vea4_thumb_thumb_thumb_thumb_thumb_t1-15.gif "Vea4_thumb_thumb_thumb_thumb_thumb_t[1]")
Note: This essay is the work of a real and still-alive dinobaby. No smart software involved, just a dumb humanoid.
Yep, alignment pipeline. No, I have zero clue what that means. I came across this felicitous phrase in “OpenAI Co-Founder Warns Superintelligent AI Must Be Controlled to Prevent Possible Human Extinction.” The “real news” story focuses on the PR push for Sam AI-Man’s OpenAI outfit. The idea for the story strikes me as a PR confection, but I am a dinobaby. Dinobabies can be skeptical.
An OpenAI professional explains to some of his friends that smart software may lead to human extinction. Maybe some dogs and cockroaches will survive. He points out that his company may save the world with an alignment pipeline. The crowd seems to be getting riled up. Someone says, “What’s an alignment pipeline.” A happy honk from the ArnoldIT logo to the ever-creative MidJourney system. (Will it be destroyed too?)
The write up reports a quote from one of Sam AI-Man’s colleagues; to wit:
“Superintelligence will be the most impactful technology humanity has ever invented, and could help us solve many of the world’s most important problems. But the vast power of superintelligence could also be very dangerous, and could lead to the disempowerment of humanity or even human extinction,” Ilya Sutskever and head of alignment Jan Leike wrote in a Tuesday blog post, saying they believe such advancements could arrive as soon as this decade.
There you go. Global warming, the threat of nuclear discharges in Japan and Ukraine, post-Covid hangover, and human extinction. Okay
What’s interesting to this dinobaby is that OpenAI made a decision to make the cloud service available. OpenAI hooked up with the thoughtful, kind, and humane Microsoft. OpenAI forced the somewhat lethargic Googzilla to shift into gear and respond.
The Murdoch article presents another OpenAI wizard output:
“Currently, we don’t have a solution for steering or controlling a potentially superintelligent AI, and preventing it from going rogue. Our current techniques for aligning AI, such as reinforcement learning from human feedback, rely on humans’ ability to supervise AI. But humans won’t be able to reliably supervise AI systems much smarter than us and so our current alignment techniques will not scale to superintelligence,” they wrote. “We need new scientific and technical breakthroughs.”
This type of jibber jabber is fascinating. I wonder why the OpenAI folks did not do a bit of that “what if” thinking before making the service available. Yeah, woulda, shoulda, coulda. It sounds to me like a driver saying to a police officer, “I didn’t mean to run over Grandma Wilson.”
How does that sound to the grand children, Grandma’s insurance company, and the judge?
Sounds good, but someone ran over Grandma Wilson, right, Mr. OpenAI wizards? Answer the question, please.
The OpenAI geniuses have an answer, and I quote:
To solve these problems, within a period of four years, they said they’re leading a new team and dedicating 20% of the compute power secured to date to this effort. “While this is an incredibly ambitious goal and we’re not guaranteed to succeed, we are optimistic that a focused, concerted effort can solve this problem,” they said.
Now the capstone:
Its goal is to devise a roughly human-level automated alignment researcher, using vast amounts of compute to scale it and “iteratively align superintelligence.” In order to do so, OpenAI will develop a scalable training method, validate the resulting model and then stress test its alignment pipeline.
Yes, the alignment pipeline. What a crock of high school science club yip yap. Par for the course today. Nice thinking, PR people. One final thought: Grandma is dead. CYA words may not impress some people. To a high school science club type, the logic and the committee make perfect sense. Good work, Mr. AI-Men.
Stephen E Arnold, July 12, 2023
-
- Subscribe to Beyond Search
Feature archive
News archive
-
