An internal fight over whether Google built technology with human-like consciousness has spilled into the open, exposing the ambitions and risks inherent in artificial intelligence that can feel all too real.
The Silicon Valley giant suspended one of its engineers last week who argued the firm’s AI system LaMDA seemed “sentient,” a claim Google officially disagrees with.
Several experts told AFP they were also highly skeptical of the consciousness claim, but said human nature and ambition could easily confuse the issue.
“The problem is that… when we encounter strings of words that belong to the languages we speak, we make sense of them,” said Emily M. Bender, a linguistics professor at University of Washington.
“We are doing the work of imagining a mind that’s not there,” she added.
LaMDA is a massively powerful system that uses advanced models and training on over 1.5 trillion words to be able to mimic how people communicate in written chats.
The system was built on a model that observes how words relate to one another and then predicts what words it thinks will come next in a sentence or paragraph, according to Google’s explanation.
“It’s still at some level just pattern matching,” said Shashank Srivastava, an assistant professor in computer science at the University of North Carolina at Chapel Hill.
“Sure you can find some strands of really what would appear meaningful conversation, some very creative text that they could generate. But it quickly devolves in many cases,” he added.
Still, assigning consciousness gets tricky.
It has often involved benchmarks like the Turing test, which a machine is considered to have passed if a human has a written chat with one, but can’t tell.
“That’s actually a fairly easy test for any AI of our vintage here in 2022 to pass,” said Mark Kingwell, a University of Toronto philosophy professor.
“A tougher test is a contextual test, the kind of thing that current systems seem to get tripped up by, common sense knowledge or background ideas — the kinds of things that algorithms have a hard time with,” he added.
– ‘No easy answers’ –
AI remains a delicate topic in and outside the tech world, one that can prompt amazement but also a bit of discomfort.
Google, in a statement, was swift and firm in downplaying whether LaMDA is self-aware.
“These systems imitate the types of exchanges found in millions of sentences, and can riff on any fantastical topic,” the company said.
“Hundreds of researchers and engineers have conversed with LaMDA and we are not aware of anyone else making… wide-ranging assertions, or anthropomorphizing LaMDA,” it added.
At least some experts viewed Google’s response as an effort to shut down the conversation on an important topic.
“I think public discussion of the issue is extremely important, because public understanding of how vexing the issue is, is key,” said academic Susan Schneider.
“There are no easy answers to questions of consciousness in machines,” added the founding director of the Center for the Future of the Mind at Florida Atlantic University.
Lack of skepticism by those working on the topic is also possible at a time when people are “swimming in a tremendous amount of AI hype,” as linguistics professor Bender put it.
“And lots and lots of money is getting thrown at this. So the people working on it have this very strong signal that they’re doing something important and real” resulting in them not necessarily “maintaining appropriate skepticism,” she added.
In recent years AI has also suffered from bad decisions — Bender cited research that found a language model could pick up racist and anti-immigrant biases from doing training on the internet.
Kingwell, the University of Toronto professor, said the question of AI sentiency is part “Brave New World” and part “1984,” two dystopian works that touch on issues like technology and human freedom.
“I think for a lot of people, they don’t really know which way to turn, and hence the anxiety,” he added.
1 in 5 companies founded in 2021 closed within the year—a story all too familiar in the US
Whether a startup is successful in its first year depends on a variety of factors—from industry type and location to funding and money management strategies. PlanPros investigated what it takes for a business to make it through its first year—a milestone that 1 in 5 companies don’t achieve.
Entrepreneurship is a core tenet of American culture. As many as 55% of Americans have started at least one business in their lifetime, according to a 2019 survey by the Global Entrepreneurship Monitor consortium at Babson College. In fact, there are over 33 million small businesses—which have fewer than 500 employees—in operation today according to estimates from the Small Business Administration. However, the Bureau of Labor Statistics reports that since 1994, about 20% of new businesses have not survived their first year.
The success of a small business affects more than just the business owners’ livelihood. According to the SBA Small Business Facts Report, small businesses are responsible for 2 in 3 jobs created in the past 25 years. Additionally, the SBA estimates that small businesses are responsible for about 44% of all economic activity in the United States.
According to a 2022 Skynova survey of 492 startup founders, 58% said they wished they had done more market research before starting their business. Put simply, market research involves evaluating how likely a product or service is to be received well by its intended customers.
Where a startup is based can have a significant effect on its finances. Business taxes vary across states, as does the availability of various government grant and loan programs designed to aid small businesses. Residents’ purchasing power also ranges geographically. The first-year failure rate for small businesses by state ranged from 18.2% to 36.6% in 2019, the most recent data available—California had the lowest first-year failure rate, while Washington-based startups faced the highest first-year failure rate.
Startups can face certain advantages and disadvantages depending on the nature of their industry as well. According to the Small Business Funding lending agency, small businesses in the health care industry have the highest chance of surviving to at least their fifth year at 60%. Conversely, small businesses in the transportation industry have the lowest chance of surviving through their fifth year at 30%.
Funding and well-managed cash flow
The primary reason new businesses fail is due to a lack of cash or available financial support in its absence, according to the aforementioned Skynova report. In 2022, 47% of startup failures were attributed to a lack of financing or investors, while running out of money contributed to 44% of failures in the same year. A 2019 study funded by the SBA of 1,000 startup small business owners attributes 82% of startup failures to cash flow problems and mismanagement. These data point out the importance of adhering to a strict budget and limiting expenses as much as possible in the first year.
It is also important to identify potential sources of funding or support in advance of any immediate need. This can help prevent running into unsustainable growth. Many government programs exist to help startups survive, including state and federal grants, some of which are designated for certain demographics and industries.
Even after a business is fairly well established, it is important to monitor cash flow closely. Businesses need to survive well beyond just the first year. According to data from the Bureau of Labor Statistics, roughly half of small businesses fail within five years. After 15 years, about 3 in 4 small businesses will have failed.
But the end of a company is not necessarily the end of entrepreneurship for every small business owner. A study by University of Michigan and Stanford economists suggests that business owners who start a second business after their first failures are more likely to succeed on their second attempt.
Story editing by Jeff Inglis. Copy editing by Tim Bruns.
Consumer retail spending holds steady as recession worries drag on
Spending for in-person and online goods and services has moderated over the past year after seeing outsized growth during the early years of the COVID-19 pandemic.
The decline in sales growth has forced some retailers to roll out discounts. Others warn sales could drop further as consumers feel the squeeze of college debt payments returning this fall and still-rising prices for everything from weekly groceries to back-to-school clothing.
Retail sales data can offer broad inferences about the spending habits of consumers in the U.S. economy and can serve as an indicator of economic health. Business leaders and Federal Reserve officials watch the data closely for signs that consumers could be struggling with their finances.
The COVID-19 pandemic and job loss caused the economy to contract rapidly and enter a recession, briefly hurting retail sales. With the help of stimulus checks and enhanced unemployment payments, Americans bounced back—and bought a ton of stuff: sporting goods, electronics, furniture, and new homes. And as COVID-19 vaccines rolled out, consumers shifted to spending on previously delayed travel.
Throughout 2021, a flood of stimulus money and rapidly rising profits contributed to a red-hot economy in which prices were rising faster than at any time since the 1980s. By 2022, Federal Reserve officials began raising interest rates in an attempt to cool down rising prices. So far in 2023, growth in sales has started to slow as consumers muddle through an increasingly expensive world.
Dom DiFurio // Shopdog
Pandemic boom in the rearview as inflation erodes spending power
Census data shows retail sales growth has slowed since federal officials began taking their fight against inflation seriously last summer.
Overall inflation was still elevated at 3.2% year over year in July, stubbornly higher than the 2% goal Federal Reserve officials want to achieve. The Fed began raising its benchmark interest rates in April 2022 to make it more expensive for consumers and businesses to borrow money for things like new business, a home mortgage, or a new vehicle.
When the Fed raises its interest rates, banks follow. A monetary policy like this aims to slow down an economy flush with cash so prices grow slower. It’s a policy experts argue causes necessary pain for consumers in the short term to avoid perpetual inflation in the long term.
So far, the Fed has seen inflation cool from a modern high of 9% last year, but in late August, officials said they need to see it decrease further before they pause rate hikes.
Dom DiFurio // Shopdog
Online retail spending is up 7.5% over summer 2022
Even as overall spending flattens, Americans’ online shopping habits have reverted to pre-pandemic trends. Consumer spending has steadily shifted from retail to e-commerce since Jeff Bezos had the idea to sell books online in 1994. That steady growth got a big boost in 2020 when online shopping often became the only way to spend money.
After correcting downward as shoppers ventured out of their homes, the portion of retail sales happening online is growing again at pre-pandemic rates despite the overall challenges faced by consumers.
However, signs are emerging from retailers that the typical American may be unable to keep spending on goods and services the same way in the second half of 2023.
QualityHD // Shutterstock
Stalled spending growth is cold comfort for nervous small retailers
Retail giant Walmart is raising its profit expectation through the end of the year, reporting gains in e-commerce sales. However, their projections starkly contrast with others in the retail space that are beginning to report gloomy forecasts.
Big sportswear retailers such as Dick’s Sporting Goods, which usually benefit from a surge in sales during back-to-school season, have cut their profit expectations for the rest of the year.
And optimism is low, according to a 2022 industry report by the National Federation of Independent Businesses, which represents the interests of America’s small businesses. Most small businesses believe the country is already in a recession despite no official call by the National Bureau of Economic Research, and they expect lower sales through the end of the year.
Story editing by Ashleigh Graf. Copy editing by Kristen Wegrzyn.
This story originally appeared on Shopdog and was produced and
distributed in partnership with Stacker Studio.
The county in every state with the most new business applications
Entrepreneurship emerged in a big way over the past three years, and founders are not slowing down.
To get a more comprehensive view of where new businesses may have the most impact in coming years, PlanBuildr used Census Bureau data to find the county in each state with the most business applications per capita in 2022.
Business applications have fallen from highs in 2021 but remain well above pre-COVID-19 pandemic levels at over 400,000 applications per month. There were a little under 5.1 million business applications in 2022, compared to a record-breaking 5.4 million in 2021 and 4.4 million in 2020. Through June, there have been 2.65 million applications, meaning 2023 is on pace to rival the 2021 spike.
Prior to the pandemic, entrepreneurship had been in a lull for decades. In 2018 and 2019, a typical month would see fewer than 300,000 business applications, and in prior years the levels were even lower, Bureau of Labor Statistics data shows. Now, entrepreneurs aren’t letting fears of a recession stop them. In July 2023 alone, more than 469,000 people applied for employer identification numbers—the primary way the government measures small business applicants—marking a 0.5% increase from June.
Additionally, small businesses and startups continue to grow jobs, increasing overall employment despite high-profile layoffs at larger corporations. Nearly half of all workers have jobs at small firms, so continued entrepreneurship keeps the job market strong. What’s more, the pioneering spirit keeps overall economic productivity strong.
Business owners told the New York Times their experiences in the pandemic had “recession-proofed” their businesses. Now they know how to pivot and survive while financially strapped. But interest rates are up, and investors have pulled back, raising the stakes for new business owners.
Groundbreaking companies, like Uber and Airbnb, have emerged from past recessions. While consumers may be more hesitant, hiring is typically easier, and business costs are less expensive amid downturns. From offering new products to creating jobs to upholding economies, new businesses create ripples across their communities.
Read on to see which county in your state saw the most businesses set the groundwork to build their legacies. The data used on new business applications are tallied from applications for employer identification numbers from the Census Bureau. States and counties that allow business owners and operators to incorporate without residing locally may have inflated per capita numbers. Counties with fewer than 1,000 residents were excluded from the analysis.
Alabama: Mobile County
– 2022 applications: 9,792 (23.8 per 1,000 residents)
– 2021 applications: 8,782 (21.3 per 1,000 residents)
Alaska: Skagway Municipality
– 2022 applications: 29 (26.8 per 1,000 residents)
– 2021 applications: 10 (8.8 per 1,000 residents)
Sean Pavone // Shutterstock
Arizona: Maricopa County
– 2022 applications: 83,305 (18.3 per 1,000 residents)
– 2021 applications: 83,458 (18.6 per 1,000 residents)
Arkansas: Phillips County
– 2022 applications: 328 (21.4 per 1,000 residents)
– 2021 applications: 347 (21.9 per 1,000 residents)
California: Alpine County
– 2022 applications: 21 (17.6 per 1,000 residents)
– 2021 applications: 23 (18.6 per 1,000 residents)
Colorado: Pitkin County
– 2022 applications: 671 (39.8 per 1,000 residents)
– 2021 applications: 727 (41.9 per 1,000 residents)
Connecticut: Western Connecticut Planning Region
– 2022 applications: 11,118 (17.8 per 1,000 residents)
– 2021 applications: Not available: Prior to 2022, data were collected for Connecticut’s former counties rather than its new planning regions. (0.0 per 1,000 residents)
Nagel Photography // Shutterstock
Delaware: Kent County
– 2022 applications: 12,961 (69.3 per 1,000 residents)
– 2021 applications: 11,552 (62.7 per 1,000 residents)
Florida: Miami-Dade County
– 2022 applications: 127,895 (47.8 per 1,000 residents)
– 2021 applications: 136,137 (51.0 per 1,000 residents)
Brett Barnhill // Shutterstock
Georgia: Fulton County
– 2022 applications: 50,118 (46.6 per 1,000 residents)
– 2021 applications: 60,986 (57.4 per 1,000 residents)
pikappa51 // Shutterstock
Hawaii: Maui County
– 2022 applications: 2,778 (16.9 per 1,000 residents)
– 2021 applications: 2,935 (17.8 per 1,000 residents)
Idaho: Teton County
– 2022 applications: 298 (23.8 per 1,000 residents)
– 2021 applications: 308 (25.1 per 1,000 residents)
Illinois: Cook County
– 2022 applications: 93,690 (18.3 per 1,000 residents)
– 2021 applications: 118,523 (22.9 per 1,000 residents)
Sean Pavone // Shutterstock
Indiana: Marion County
– 2022 applications: 22,226 (22.9 per 1,000 residents)
– 2021 applications: 25,352 (26.1 per 1,000 residents)
stivanderson // Shutterstock
Iowa: Jefferson County
– 2022 applications: 221 (14.1 per 1,000 residents)
– 2021 applications: 225 (14.3 per 1,000 residents)
TommyBrison // Shutterstock
Kansas: Lane County
– 2022 applications: 45 (28.9 per 1,000 residents)
– 2021 applications: 22 (14.0 per 1,000 residents)
f11photo // Shutterstock
Kentucky: Jefferson County
– 2022 applications: 12,082 (15.6 per 1,000 residents)
– 2021 applications: 12,973 (16.7 per 1,000 residents)
Louisiana: Orleans Parish
– 2022 applications: 9,408 (25.4 per 1,000 residents)
– 2021 applications: 12,809 (34.0 per 1,000 residents)
Joseph Sohm // Shutterstock
Maine: Cumberland County
– 2022 applications: 3,670 (11.9 per 1,000 residents)
– 2021 applications: 3,803 (12.4 per 1,000 residents)
Maryland: Garrett County
– 2022 applications: 748 (26.2 per 1,000 residents)
– 2021 applications: 579 (20.1 per 1,000 residents)
Massachusetts: Nantucket County
– 2022 applications: 433 (30.0 per 1,000 residents)
– 2021 applications: 344 (23.7 per 1,000 residents)
Michigan: Wayne County
– 2022 applications: 39,328 (22.4 per 1,000 residents)
– 2021 applications: 47,565 (26.8 per 1,000 residents)
Minnesota: Cook County
– 2022 applications: 94 (16.5 per 1,000 residents)
– 2021 applications: 81 (14.4 per 1,000 residents)
Mississippi: Coahoma County
– 2022 applications: 774 (38.3 per 1,000 residents)
– 2021 applications: 754 (36.4 per 1,000 residents)
Sean Pavone // Shutterstock
Missouri: St. Louis city
– 2022 applications: 6,422 (22.4 per 1,000 residents)
– 2021 applications: 7,803 (26.6 per 1,000 residents)
Montana: Flathead County
– 2022 applications: 4,219 (37.7 per 1,000 residents)
– 2021 applications: 3,521 (32.4 per 1,000 residents)
Nebraska: Boyd County
– 2022 applications: 33 (19.0 per 1,000 residents)
– 2021 applications: 18 (10.1 per 1,000 residents)
randy andy // Shutterstock
Nevada: Clark County
– 2022 applications: 49,369 (21.3 per 1,000 residents)
– 2021 applications: 55,475 (24.2 per 1,000 residents)
Wangkun Jia // Shutterstock
New Hampshire: Rockingham County
– 2022 applications: 3,582 (11.2 per 1,000 residents)
– 2021 applications: 3,559 (11.2 per 1,000 residents)
Mihai_Andritoiu // Shutterstock
New Jersey: Essex County
– 2022 applications: 18,114 (21.3 per 1,000 residents)
– 2021 applications: 21,329 (25.0 per 1,000 residents)
Jimack // Shutterstock
New Mexico: Santa Fe County
– 2022 applications: 4,338 (27.9 per 1,000 residents)
– 2021 applications: 3,683 (23.7 per 1,000 residents)
pisaphotography // Shutterstock
New York: New York County
– 2022 applications: 50,149 (31.4 per 1,000 residents)
– 2021 applications: 53,217 (33.7 per 1,000 residents)
digidreamgrafix // Shutterstock
North Carolina: Mecklenburg County
– 2022 applications: 29,600 (25.8 per 1,000 residents)
– 2021 applications: 34,052 (30.2 per 1,000 residents)
Traveller70 // Shutterstock
North Dakota: Dunn County
– 2022 applications: 88 (21.9 per 1,000 residents)
– 2021 applications: 62 (15.4 per 1,000 residents)
Sean Pavone // Shutterstock
Ohio: Franklin County
– 2022 applications: 24,370 (18.4 per 1,000 residents)
– 2021 applications: 26,733 (20.3 per 1,000 residents)
Oklahoma: Oklahoma County
– 2022 applications: 14,955 (18.6 per 1,000 residents)
– 2021 applications: 16,158 (20.2 per 1,000 residents)
Oregon: Hood River County
– 2022 applications: 409 (17.0 per 1,000 residents)
– 2021 applications: 352 (14.6 per 1,000 residents)
Pennsylvania: Philadelphia County
– 2022 applications: 29,166 (18.6 per 1,000 residents)
– 2021 applications: 42,298 (26.6 per 1,000 residents)
Rhode Island: Providence County
– 2022 applications: 7,055 (10.7 per 1,000 residents)
– 2021 applications: 7,450 (11.3 per 1,000 residents)
South Carolina: Charleston County
– 2022 applications: 10,138 (24.2 per 1,000 residents)
– 2021 applications: 10,561 (25.5 per 1,000 residents)
South Dakota: Haakon County
– 2022 applications: 37 (20.3 per 1,000 residents)
– 2021 applications: 19 (10.4 per 1,000 residents)
Tennessee: Davidson County
– 2022 applications: 15,302 (21.6 per 1,000 residents)
– 2021 applications: 15,871 (22.6 per 1,000 residents)
Texas: Glasscock County
– 2022 applications: 30 (25.8 per 1,000 residents)
– 2021 applications: 13 (11.4 per 1,000 residents)
Utah: Summit County
– 2022 applications: 1,291 (30.0 per 1,000 residents)
– 2021 applications: 1,457 (33.8 per 1,000 residents)
Joseph Sohm // Shutterstock
Vermont: Lamoille County
– 2022 applications: 374 (14.3 per 1,000 residents)
– 2021 applications: 322 (12.3 per 1,000 residents)
OJUP // Shutterstock
Virginia: Petersburg city
– 2022 applications: 797 (23.9 per 1,000 residents)
– 2021 applications: 1,017 (30.5 per 1,000 residents)
Washington: San Juan County
– 2022 applications: 273 (14.6 per 1,000 residents)
– 2021 applications: 356 (19.1 per 1,000 residents)
West Virginia: Jefferson County
– 2022 applications: 646 (11.0 per 1,000 residents)
– 2021 applications: 569 (9.7 per 1,000 residents)
Wisconsin: Milwaukee County
– 2022 applications: 18,019 (19.6 per 1,000 residents)
– 2021 applications: 19,542 (21.1 per 1,000 residents)
Ems Images // Shutterstock
Wyoming: Sheridan County
– 2022 applications: 22,389 (697.6 per 1,000 residents)
– 2021 applications: 17,043 (538.2 per 1,000 residents)
Data reporting by Paxtyn Merten. Story editing by Jeff Inglis. Copy editing by Paris Close.
This story originally appeared on PlanBuildr and was produced and
distributed in partnership with Stacker Studio.
Events2 months ago
Where will AI go next?￼
Business4 months ago
How to build company culture in a scale-up
Events5 months ago
The innovator’s mindset and the battle between Batman-v-Superman: mesh conference day 2
Business3 months ago
How to build and maintain a company culture among a remote workforce
Technology6 months ago
How are AI tools like ChatGPT deployed in retail?