Scientists Discover Powerful Antibiotic Using AI

The beneficial power of AI. JL

The BBC reports:

Scientists have discovered a new type of antibiotic using artificial intelligence (AI).A powerful algorithm was used to analyse more than one hundred million chemical compounds in a matter of days. Scientists trained it to analyse the structure of 2,500 drugs and other compounds to find those with the most anti-bacterial qualities that could kill E. coli. They then selected around 100 candidates for physical testing before discovering halicin. The newly discovered compound was able to kill 35 types of potentially deadly bacteria.

A Nation of Voyeurs: How Ring And Nest Normalized Surveillance

Is the marketing of a device to make people feel safer actually making them more fearful and paranoid? JL
 
Drew Harwell reports in the Washington Post:

In-home and outdoor camera systems across the United States have reshaped daily lives. They analyze neighbors, monitor kids and house guests, judge housekeepers and babysitters. The motion-activated devices send alerts whenever someone walks by and can be triggered by cars, dogs, squirrels and windblown trees, leading some customers to feel under siege. Amazon engineer Max Eliaser wrote last month that the mass deployment of Internet-connected cameras was “not compatible with a free society. The privacy issues are not fixable with regulation, and there is no balance that can be struck.”

YouTube Leads By So Much In Mobile Streaming It's Not Even Close

It's content tends to be shorter than that on other entertainment apps and the crossover appeal of music videos helps. Oh, and it's free.

So the strategic question is how Netflix, Disney et al can compete? JL 

Julia Alexander reports in The Verge

In 2019, YouTube dominated 70% of the total time people spent on their phones watching the top five entertainment apps where more people are spending their time, especially teenagers and young adults. The crossover between YouTube and YouTube Music helps keep people in YouTube’s ecosystem. “YouTube is seven times larger than (Netflix) in viewing hours. Of course, it’s free. So the real question is, can (Netflix) produce enough content people are willing to pay for?” 

Bioprinting - The 3D Printing Of Body Part Tissue - Is Here

Prosthetics were the functional equivalent of training wheels. Bioprinting techniques are creating cells and heart valves. JL


Dinusha Mendis and Ana Rutschman report in Singularity Hub:

Medical professionals now routinely 3D print prosthetic hands and surgical tools. Bioprinting uses 3D printers and techniques to fabricate the three-dimensional structures of biological materials, from cells to biochemicals, through precise layer-by-layer positioning. The ultimate goal is to replicate functioning tissue and material, such as organs, which can then be transplanted into human beings. Human heart valves are now being 3D printed although none have been transplanted into people yet.

Uber And Lyft Said They'd Reduce Congestion. Data Show They've Made It Worse

A growing body of data suggests tech generally - and gig economy 'benefits' to society specifically - have created unintended consequences worse than the problem they were ostensibly designed to solve.

In addition to ride hailing, there is Airbnb's impact on housing, delivery apps' impact on restaurant workers' wages and Facebook's impact on politics. JL

Eliot Brown reports in the Wall Street Journal:

A mounting collection of studies show that far from easing traffic, Uber and Lyft are adding to congestion in U.S. downtowns. Rather than the apps (introducing) algorithm-driven efficiency, drivers in major cities cruise for fares without passengers 40% of the time. Uber and Lyft have pulled people away from buses, subways and walking, and the apps add to the amount of driving in the U.S. The reversal of ride-hailing from hero to villain is the sort of unintended consequence that has become a recurring feature of Silicon Valley disruption.

The Coronavirus Has Cut China's Co2 Emissions By 25 Percent

Of course, exposing your population to a deadly pandemic while shutting down industrial capacity is a rather extreme way to reduce emissions. JL

Lauri Myllyvirta reports in Carbon Brief:

The measures to contain coronavirus have resulted in reductions of 15% to 40% in output across key industrial sectors. This is likely to have wiped out a quarter or more of the country’s CO2 emissions over the past two weeks, the period when activity would normally have resumed after the Chinese new-year holiday. Reductions in coal and crude oil use indicate a reduction in CO2 emissions of 25%. (But) shaving 25% off energy consumption and emissions for two weeks would only reduce annual figures by 1%.

Pitfalls Of the Connected Rent-A-Car

About the lack of internet, cell or wifi access...JL

Kari Paul reports in The Guardian:

A growing field of car sharing services make it incredibly convenient to hit the road – no paperwork or trips to the rental company required. Cars are re-synced automatically every 24 hours. The car needs to update its connection to the cloud for security purposes. (But) without the ability to re-sync the car, customer service agents cannot remotely assist members. The car cannot be re-synced if it is more than 50 miles outside of the “HomeZone”, out of cell range, or if the vehicle is moving. “It works great if you have a dependable cell connection.”

How Grocers Are Using Data To Wrest Control of Shelf Space From Food Brands

Shedding reliance on brands in the co-evolutionary struggle over data-driven dominance, grocery retailers are now employing software and algorithms that give them more information - and control - over where, when and how to present products on shelves. JL


Annie Gasparo and Jaewon Kang report in the Wall Street Journal:

Grocers are relying on proprietary research to decide where to shelve products, rather than relying on companies that sell well-known brands to tell them what to put on what shelf at what price. Retailers are using sophisticated software to decide where to place items, factors that can move sales several percentage points. Algorithms measure how much time a customer is willing to spend looking for products, resulting in less space for traditional staples in favor of niche items and store brands that deliver higher margins and higher demand. "Retailers have better information now.”

To Understand the Future Of Work, Look At Transcription

Rather than replacing workers, technology is creating more jobs, but they are more routine - and far more likely to be poorly paid contract labor. JL 

Clive Thompson reports in Wired:

Demand for transcription has actually exploded in recent years. Why? Because audio is easier than ever to capture (via our pocket computers), so people are recording ever more meetings. Plus, video and podcasting have become the dominant forms of rhetoric. Daily communication is increasingly multimedia. But we still can't search the contents of video or audio well, so we need to transcribe it. Theoretically, exploding demand would drive up the price of labor. Except globalization and the gig business model exploded the supply of workers. AI doesnt always destroy jobs. Just makes them more likely to suck. 

Why 'You're Not Listening' - And Technology Doesn't Help

It's the most connected age in human history; more people that ever are communicating with those they know - and perhaps even more with those they don't. But at the same time, loneliness has become a mental health epidemic in developed countries, raising questions about the unintended consequences of such social engineering.

It may be that the same devices are enabling both trends, making it easier to 'curate'  those communications, creating distance through the substitution of symbols like emojis and short bursts of text for the slower, more intense art of conversation. And that becoming a 'generous listener' is the rarest gift humans can now bestow. JL

Kate Murphy reports in the New York Times:

In a survey of 20,000 Americans, half said they did not have meaningful in-person social interactions, such as an extended conversation with a friend, on a daily basis. The same proportion said they felt isolated. A lack of listening is a contributor to feelings of loneliness. Technology doesn’t help. Devices are a constant distraction, and people tend to be inaccurate at interpreting feelings through text and emoji. The best way to understand those closest to us is to spend time with them, put down phones and actually listen to what they have to say.

Google Cloud AI Is Removing Gender Identity Labels From People In Its Images

An attempt to reduce bias, which may improve performance verses competitors - and which may also be a clever marketing move. JL

Khari Johnson reports in Venture Beat:

Google Cloud AI is removing the ability to label people in images as “man” or “woman” with its Cloud Vision API (which) provides computer vision for customers to detect objects and faces.“Given that a person’s gender cannot be inferred by appearance, we have decided to remove these labels in order to align with the artificial intelligence principles at Google, specifically to avoid creating or reinforcing unfair bias.” Researchers found that AI from Amazon, Clarifai, Microsoft, and others misidentified people with no gender identity 100% of the time.

Amazon, Facing Merchant Defections, Tweaks Algorithmic Recommendations

As merchants become more technologically sophisticated they are learning how little - or much - they can get from partnering with Amazon. And as more of them determine to what degree Amazon sublimated their brands to Amazon's own strategic imperatives, they are going their own way.

To counter this nascent trend, Amazon is beginning to promote brands on special digital 'storefronts' where merchants don't have to fear that their products will be undercut by rivals - or by Amazon itself. We'll see how long that lasts. JL

Seb Joseph reports in Digiday:

Marketers have for years had problems with Amazon’s search recommendation engine, an algorithm that was responsible for over 35% of all sales on Amazon.com in 2018. As larger, premium companies establish their own direct-to-consumer offering, more (wonder) whether they can work with Amazon when they’ve had limited influence over what is recommended to shoppers. Amazon is hoping storefronts, a branded site within Amazon.com where marketers can showcase products without having to worry about shoppers being recommended to buy similar items from a rival, will assuage concerns.

The Five Industries That Currently Rely the Most On AI

It is noteworthy that the five industries are disparate in their use of AI - and may be quite different from those in the future as AI evolves and usage becomes more sophisticated.

But ultimately, this is all about interpreting data and using it to diagnose future behavior. JL

Analytics Insight reports:

AI has made online advertising more effective than traditional marketing tools. Online casinos rely heavily on AI; if they didn’t have this technology, they wouldn’t exist. Online casinos use AI to enforce fair-play by making every outcome of every game random. In healthcare, AI helps doctors by providing them with better diagnostics and detecting a medical problem much faster. AI is playing a key role in manufacturing by making the process more efficient. In transportation, AI has the potential to (determine) the future.

The Government Uses Facial Recognition Because It Doesnt Trust You

And it thinks it has good reason to do so. JL


Matthew Guariglia reports in Slate:

Suspicion is a circular process.Today a single encounter with the government that involves a photograph—whether for a driver’s license or a mug shot—has the chance to beget more. Technological advances have meant the moral quandary of the rogues’ gallery is a supersize version of what it once was. People whose faces hang in the police station can be denied jobs or passports because of misreadings or misunderstandings buried in hoarded data. Face recognition means we are all constantly under suspicion. It means your government doesn’t trust you.

The Reason Americans Should Be Furious About Their Internet Service - And Bills

The US doesnt have an internet or broad band problem, it has an oligopoly problem. JL

Emily Stewart reports in Re/code:

The US has fallen behind developed economies in broadband penetration. Prices are significantly higher. The average monthly cost of broadband in America was $66.17; in France,  $38.10, in Germany, $35.71, and in South Korea, $29.90. Corporate concentration means companies have to compete less for workers pushing wages down. Oligopolies also harm suppliers. And it’s hardly just the internet. Similar phenomena (exist) in cellphone plans, airline prices, and other arenas, due to a lack of competition. Corporate consolidation is costing American households an extra $5,000 a year.

Research: American Manufacturing's Resurgence Is A Myth

The reality is that manufacturing has grown less than the rest of the economy and has begun shedding jobs.

Despite the benefit of tax cuts, the trade war and related policies have negatively impacted performance, including profits, employment and compensation, to such a degree that not even automation can save it at this point in the business cycle. JL


Joe Ragazzo reports in Talking Points Memo:

Manufacturing entered a recession in 2019. By October of last year, U.S. manufacturing had seen two consecutive quarters of contraction. The sector shed 5,000 jobs in December and 12,000 jobs in January. In December, the manufacturing index displayed the fastest rate of contraction since June 2009. Manufacturing employment is 34% lower than it was at its peak in 1979. Manufacturing employment has grown 3.5% since (2017), lagging overall employment which has grown 4.5%. By 2018, manufacturing workers were earning 5% less than their private sector counterparts.

Why Employees Are Increasingly Demanding Escape From Open Offices' Din

The open plan office made even traditional companies look like cool startups. And they saved money on rising real estate costs.

But as data has proliferated on the negative impact such designs have on individual performance, organizations are adapting by installing small spaces which employees can use when they need to concentrate - or have an actual conversation. JL


Jane Margolies reports in the New York Times:

70% of offices in the United States have open plan offices. Such plans took off in the name of teamwork but face-to-face interactions fell 70% when firms switched to open offices. Confidential conversations became impossible in a sea of colleagues.A decade since the swing toward open-plan, and the resulting backlash from workers concerned about noise and a lack of privacy, ancillary spaces are cropping up, offering employees an escape from co-workers. These include prayer rooms, wellness rooms, libraries, phone booths, mini meeting rooms and pods.

Could AI Really Assist With the Growing Demand For Mental Health Treatment?

The global growth rate for mental health disorders, many of them untreated due to cost or lack of available professional help, is astounding. JL

Jonathan Burton reports in Market Watch:

As mental disorders rise — the cost to the global economy is projected to be $16 trillion over the next decade, caring for patients with precision is a Holy Grail for mental-health professionals. Current diagnosis and treatment methods, while skilled and insightful, cannot fully capture the unique needs and complexity of every patient — not without time, money and a willingness that many people do not have. AI-based therapies have the potential to be faster and cheaper, which can encourage patients to continue their counseling.

US Military Reveals AI Facial Recognition That Identifies Subject From A Kilometer Away

 So much for 'wish you were here...' JL

Luke Dormehl reports in Digital Trends:

The United States military is in the process of funding the creation of a portable face-recognition device that’s able to identify individuals from up to 1 kilometer away. Even with cutting-edge A.I., it’s a challenging problem to solve, since adding longer lenses to cameras increases vibrations. At these distances, atmospheric turbulence can distort the image similar to the heat shimmer seen on hot days. A neural network could be used to unscramble the distorted image in order to get workable facial data for identification purposes. A working prototype of the system was demonstrated at the end of last year.

Tesla Cybertruck Pre-Orders Now Exceed 500,000

Which puts in some perspective Elon Musk's mockery of Bill Gates for buying a Porsche rather than a Tesla. JL 

The Next Web reports:

The 522,764 pre-orders represent an average of 5,873 orders received per day in the 89 days since (and including) reveal night. Out of those pre-orders, 17% are for the one-motor version of the Cybertruck, which starts at $40,000. Tesla is hoping to launch the two higher-end models by late 2021, but taking into account the company’s record of missing production targets — that seems like an optimistic goal.

Why Prediction Markets Are Bad At Predicting Who'll Be President

Current prediction markets are too small to scale - and there is not yet enough of a financial reward to encourage greater participation - which makes them prone to randomness and inaccuracy. JL


Kelsey Piper reports in Vox:

Prediction markets were supposed to be smarter than the pundits. They were supposed to harness the wisdom of crowds and use financial incentives to be as accurate at predicting global events as the stock market is at predicting earnings for public companies. Existing prediction markets are too small in scale, hard to interact with, and hard to make money from, which renders them inaccurate and vulnerable to manipulation.

The Reason Amazon Is Causing the Economy To Drown In Weird Trademarks

Millions of products sold on Amazon come from China. To get registered on Amazon, however, the seller needs a registered trademark. And it turns out that the sometimes nonsensical sounding brands from China have a much higher approval rate than do traditional applications from US companies.  

The larger strategic question is whether, in the digital age, Amazon has rendered the importance of brands obsolete compared to the power of its platform as a sales promotion tool. JL


John Herrman reports in the New York Times:

If you’re feeding a brand-new listing into Amazon and doing so without a pre-existing brand, getting into Brand Registry is extremely important. It requires a registered trademark. These tens of thousands of applications have a higher registration rate than large U.S. companies.” Their success in getting approved can likely be credited to the names: a completely novel application will just “fly through.” Other theories for the rise include monetary incentives provided by the Chinese government for obtaining foreign intellectual property rights“It’s100% due to Amazon e-commerce sellers.”

China Is Using Smartphone App Health Rating To Decide Who Can Go Back To Work

This government-big tech approach to disease control may  based on self reported smartphone apps may be inefficient because it encourages people to lie if they are infected but want to go back to work.

The cynical view may be that it is more important to get the economy up and running again than to suffer a few (thousand?) more casualties. JL


Lisa Lin reports in the Wall Street Journal:

Alibaba and Ant Financial worked with the government to develop a smartphone-based system to classify people into categories of exposure—green, yellow or red—based on their health conditions and travel history. Tencent created a similar program.Those marked green are clear to roam freely, and are given a QR code to present at checkpoints. Working with the companies, the government can tap expertise it lacks. The companies benefit from a government-mandated boost to traffic on their platforms. Travel restrictions and self-reported data could fail because the stigma incentivizes people to lie.

How AI Predictive Modeling Is Beginning To Help Leaders Make Better Decisions

Business analytics software has made leadership decision-making more productive over the past 25 years. So why isnt every organization like Amazon or Google?

What AI and its related modeling techniques provide is the ability to efficiently process the vastly expanded quantities of data available - and more importantly, encourage leaders to think more expansively and creatively about the opportunities and threats they face, enabling optimization of the information, the technology and organizational potential. JL

Ravi Bapna reports in Knowledge@Wharton:

In predictive modeling, we start with past data and outcomes as inputs, then use computation to ‘learn’ the rules that map the input data to an outcome. Descriptive analytics, a branch of machine learning, generates value by identifying patterns in hyper-dimensional data. Executives understand a several thousand dimensional dataset can detect anomalies in a hyper-dimensional space, (such as) unobservable factors that might be driving the outcome. Where combinatorial optimization algorithms excel is in optimizing goals in the presence of a variety of constraints.

Deploying Machine Learning To Manage the Tidal Wave of IoT Data

Billions of sensors are being installed to capture data on everything from trains to security surveillance cameras to smart home appliances. Increasingly, AI and ML are employed to make sense of the growing amounts of data in order to optimize the performance of all the connected devices and the information they generate.

For now, it is making the process manageable. The degree to which it will improve performance is not yet clear. JL

Smriti Srivastava reports in Analytics Insight:

Currently, there is a lot of manual input required to achieve optimal functionality; there is not a lot of intelligence built-in. Every time the IoT sensors gather data, someone (has) to classify the data, process them and ensure information is sent back to the device for decision making. The pattern-matching abilities of machine learning uses information to improve business processes, products, and services. The most common for machine learning and IoT data will be predictive maintenance, analyzing CCTV surveillance, smart home applications, in-store ‘contextualized marketing’ and intelligent transportation systems.

Google Changes Borders On Its Digital Maps Depending On Who's Looking

As technology has given tech companies the ability to show different content to different users, its maps now reflect whatever the dominant political/cultural truth in the region from which that person is logged in.

There is a lot of advertising revenue at stake, so the company does not want to miss any of it. JL

Greg Bensinger reports in the Washington Post:

With some 80% market share in mobile maps and over a billion users, Google Maps has an outsize impact on people’s perception of the world. Maps are a big business for Google, generating $3.6 billion in annual sales by next year, primarily through advertising. When it comes to contested borders, people in different countries often see different things. Google has special employees referred to as “the disputed region team” that addresses prickly matters.

Algorithms And Humans Are Equally Bad At Guessing Which Criminals Will Re-Offend

Arguments a persist as to whether computers and algorithms are better at predicting a re-occurrence of criminal behavior.

But the larger point is that whichever may slightly outperform the other depending on the test methodology or data, the fact remains that accuracy for either is only at at 65 or 70% level, which seems pretty low when deciding whether someone has to spend time in jail. JL

Sophie Bushwick reports in Scientific American:

Untrained humans performed as well as a popular risk-assessment software at forecasting recidivism, or whether a convicted criminal would reoffend. "These tools are being used in a courtroom context, where they have very profound significance—and can highly impact somebody’s life if they are held in jail for weeks before their trial. We should be holding them to a higher standard than 65 to 70% accuracy—and barely better than human predictions.”

Why Netflix Signed Exclusive Content Deal With Samsung To Battle Streaming Rivals

Electronic ecosystems are increasingly important to streaming distribution. JL

Sarah Perez reports in Tech Crunch:

The partnership comes at a critical time for Netflix. Its subscriber growth in the U.S. has gone flat against new streaming competitors including Disney, Apple , HBO , NBCU and Quibi. These new streaming services already have ways to integrate with mobile devices or partnerships allowing them to distribute their service to millions. Its only option to get similar global scale is Samsung, which had 18.8% worldwide market share. Netflix will offer bonus content to Samsung device owners that won’t be found elsewhere.

Digital Era Marketing: Customers As Tribes, Stores As Churches

As traditional institutions like religion and government have become sources of disillusionment and discord, brands are increasingly conferring a statement by the consumer about belonging, about tribal attachments and the values associated with that.

Successful companies recognize that everything from the way the organization treats its employees to the quality and price of its products must be consistent with that message. JL


Paul Hiebert reports in Ad Week:

Brands that can establish, grow and manage a sense of collective togetherness among its customer base will increase their chances of success. “The hallmarks of community are exceptional levels of engagement between members. Customers are part of tribes, and the store is a showroom or clubhouse. It’s a church to the brand. “The future is bricks plus clicks. You have to create moments that are consistent with the brand and help drive wonderment." (And) “What’s increasingly important for customers is mission and purpose. That needs to permeate the whole organization, as well as your product.”

The Enduring Dominance of California Capitalism

Despite its notorious reputation for a housing affordability crisis, apocalyptic wildfires and high taxes, California continues to lead the nation in income growth, new jobs, business formation and venture capital investment.

The disparity reflects its attractiveness as a place to live and work, thanks - in no small part - to smart public policies. JL

Laura Tyson and Lenny Mendonca report in Project Syndicate:

California, the world’s fifth-largest economy, is developing its own political economy. Since 2008, California has added two million payroll jobs, the unemployment rate has fallen to a record low of 3.9%, and average per capita income has increased 25%. Since t 2010, California has added 3.4 million jobs,15% of the nation’s total. With four million small businesses, California also leads the country in new business formation. It receives more than half of all US venture-capital funding. California continues to lead the country in environmental policies, too.

EU's New AI Rules Will Focus on Ethics and Transparency

The question is whether Europe really matters enough to the US and China for these rules to make a significant difference. JL

Chris O'Brien reports in Venture Beat:

Europe has in recent years sought to emphasize fairness and ethics when it comes to tech policy. Now it’s taking that approach a step further by introducing rules about transparency around data-gathering for technologies like AI and facial recognition. These systems would require human oversight and audits. The plan assumes Europe has a wealth of government and industrial data, and it wants to provide regulatory and financial incentives to pool that data, which would then be available to AI developers who agree to abide by EU regulations.

Analysis Shows YouTube's Algorithms Aren't Biased, But Flag Misinformation and Hate Speech

Research now shows that the algorithms themselves are ideologically neutral  and unbiased.

But, right wing complaints of political bias appear to be based on the fact that YouTube has policies  outlawing basic misinformation and hate speech, which apparently occur more in videos posted by that end of the political spectrum, making decisions to moderate them open to accusations of bias by those who post them. JL

Zain Humayun reports in ars technica:

Studies of bias are difficult because the same results could be caused by confounding variables. Right-leaning videos attract moderation because they contain erroneous information or hate speech. "There is no political censorship. YouTube appears to just be enforcing their policies against hate speech." Every decision they make to censor or allow content draws criticism from the other side of the political spectrum. "We're so heavily polarized now no one will ever be happy."

The Downside of AI Diagnosis By Smartphone

Based on tech's history with personal information, the economic imperative to use such data in abusive ways is not just possible, but probable. JL


Lois Parshley reports in Vox:

Deducing details of a person’s health through how often they text is digital phenotyping. Phenotypes are traits derived from how your genes interact with your environment. These tools may predict illnesses before they would be diagnosed. (But) insurance companies could base rates on conditions you might not be aware of.  Could educational institutions access some of this behavioral health information admission? Employers (could) filter candidates based on what apps you use. “Are there repercussions they may not be aware of because their data is out there?”

How Much Is An Oscar Worth?

Much of the value is intangible, bestowing reputational advantage on those associated with it, who can then attempt to monetize that through contracts for future projects.

And there may be more value in the nomination than the actual award, especially for an entertainment industry increasingly dependent on streaming. JL


Paul Bernstein reports in Fortune:

The Academy’s official regulations say winners cannot sell their Oscar without first offering to sell it back to the academy for $1. Decades ago, Oscar-winning movies could rely on the “Oscar bounce” at the box office. That’s no longer the case because movies stay in theaters for shorter time periods. “Nowadays, the Oscar is more of a nomination bounce. By the time of the telecast, most movies are completely played out.” The nominations are added value for streaming.

Multiple Tech Company Layoffs Shock Silicon Valley

What is not yet clear is whether this is a cyclical repositioning or a structural readjustment based on longer term trends. Either way, it is worrisome to tech and to the broader economy. JL

George Avalos reports in the San Jose Mercury News:

At least eight Bay Area tech companies have alerted the state government of upcoming job cuts that top 1,000 positions. Zume cut 252 positions in Santa Clara and San Francisco; VMWare eliminated 211 jobs in Palo Alto; Shutterfly chopped 153 positions in Redwood City; Intel cut 153 jobs in Santa Clara; Comcast eliminated 127 positions in Livermore; Xilinx shed 123 jobs in San Jose; 23andMe cut 84 jobs in Sunnyvale; and NortonLifeLock jettisoned 59 jobs in Mountain View.

Silicon Valley Continues To Lose People As Costs Rise

Silicon Valley is a famously expensive place to live. The combination of economic opportunity, celebrity, a beautiful landscape and great weather is hard to beat.

But the looming question is whether those rising costs make it so difficult for companies to attract the talent they need (and operate profitably) that they will ultimately have to relocate many of their operations in order to compete globally.  JL


Abrar Al-Heeti reports in CNET:

For the third year in a row, more people are leaving the area than are coming in. Silicon Valley is seeing employment growth and record megadeals, but they're coming alongside rising costs, especially housing and income inequality. There have been 821,000 new jobs created in the Bay Area since the recession, but only 173,000 new housing units. The median home price is more than $1 million, making it the highest in the nation. As 13% of households hold 75% of the region's wealth, 30% of residents aren't meeting self-sufficiency standards as the cost of goods and services has gone up.

Why the Behavioral Economics of Retail Failure Goes Beyond Ecommerce

Conventional wisdom blames the collapse of retail shopping on ecommerce generally and Amazon specifically. But the 'retail apocalypse' may have more to do with behavioral economics than technology.

The data reveal that consumers have changed not just where they shop, but what they shop for and how much money they have to shop. The cumulative impact of these changes in the way people behave may do more to explain the decline of main street and malls than do smartphones and websites alone. JL

Austan Goolsbee reports in the New York Times:

While ecommerce is growing, it represents only 11% of retail sales. Major economic forces have had a bigger impact on brick-and-mortar retail than the internet. 1)We have changed where we shop, away from smaller stores in malls and toward “Big Box” stores. 2) 70% of retail spending in the US is in categories that have had slow encroachment from the internet. This includes automobiles, gasoline, home and garden supplies, drugs, food and drink. 3) Since 1970, the share of income earned by the middle class has fallen from almost 66% to 40%. 4) Americans have spent less income on things and more on services.

The NBA Is Using AI To Produce Highlight Clips From Its All-Star Games

Dunk! (The annual NBA All Star game is tonight).

Jessica Golden reports in CNBC, Image courtesy of Sports Illustrated and the NBA:

As social media has emerged as important to reach fans, the need for more customized highlights has grown. The NBA has employed technology to analyze key moments of each game and the best highlights. (It) uses machine learning or AI to take a combination of visual, audio and data cues to identify big moments in games to create shareable highlights. The software will automatically create multiple clips and content for every single player. “Previously, it could take an hour to cut a post-game highlights package. Now it takes a few minutes to create 1,000 highlight packages.”

Why Pop Songs Are Trending Sadder

Just as negative emotions have become predominant in social media because of their impact on engagement, so negative lyrics may have a similar effect in pop music in this era. JL

Charlotte Brand reports in Aeon:

We found small evidence for success and prestige bias operating in the datasets. (But) content bias was the most reliable effect in explaining the rise of negative lyrics. In cultural evolution, negative information appears to be remembered and transmitted more than neutral or positive information. In 1965, 450 words were associated with negative emotions. In 2015 their number was above 700.

Even Non-Residents Of California Can Use Its Law To Stop Sale Of Their Data

Which is great, though the question about its impact will depend on how many people avail themselves of the right to do so. JL

Geoffrey Fowler reports in the Washington Post:

Americans living anywhere can use the CCPA to reset their relationships with more than a dozen major businesses including Netflix, Microsoft, Starbucks and UPS. You can ask companies to show you exactly what data they’ve collected about you.You can instruct companies not to “sell” your data. The word “sell” is in quotes because the law defines that broadly as an exchange of value. (There’s a lot of debate about that). You can ask companies to delete your data, unless doing so would create a security threat or interfere with someone else’s free speech.

How Google Got Its Employees To Eat Their Vegetables

Changing not just the food, but how it's presented and where it's located. JL

Jane Black reports in One Zero:

The company has taken a Google-ish approach to food - methodical, iterative - to create the largest real-world test of how to make healthier choices at mealtime. The campaign isn’t changing just the food itself, but how it’s presented. Google’s limiting portion sizes for meat and desserts and redesigning its premises to lead “users” to choose water and fruit over soda and M&M’s. Seafood consumption jumped 85% between 2017 and 2018. In 2018, Googlers drank five times more water than sugary drinks. Instead of the usual 6.5 feet from coffee, the snack table was placed 17 feet away. That four or five extra steps reduced snacking by 23% for men and 17% for women.

Thy Neighbors' Solar Panels

The answer to increasing environmentally beneficial behavior may be as or more dependent on psychology than technology or government policy. JL


Robert Frank reports in The Atlantic:

Social scientists have demonstrated the influence of peer behavior. Behavioral contagion can exacerbate harmful behaviors. We’ve been engaging in energy-intensive activities because of our tendency to behave as our peers do. (But) where contagion creates a problem, it can also solve it. Solar-panel adoption is particularly contagious. Each new installation in a neighborhood can lead to several additional ones. (solar panels visible from the street exert greater peer effect than those that aren’t). Installing solar panels or buying an electric vehicle is likely to spur others to take similar steps.

Costco Capitalism's Winning Formula

As consumer suspicion and anger grows, Costco may be the avatar of the post-marketing environment, in which pricing and perceptions of fairness in customer and employee treatment enable an immensely profitable business model. JL


Bryan Lehrer reports in his blog:

Costco is #2 in the world in the retail sector. It arrives at low prices through psychologically favorable tactics for stakeholders. Walmart earns $235,450 of sales per employee, Costco earns $744,893. This means they can increase employee wages, which increases talent and retention levels, which increases customer happiness, which increases sales. The efficiency enabled by selling in bulk, in a warehouse means there are no distribution center workers rights infringements, no data capture scandals. Costco finds success playing along with the rules fairly, not bending them to their will.