Connect with us

Business

Deloitte allegedly cited AI-generated research in a million-dollar report for a Canadian provincial government

Published

on



A Canadian government-commissioned Deloitte health care report that cost one province nearly $1.6 million contains potentially AI-generated errors, marking the second country this year to allege the consulting firm’s fact-checking shortcomings. 

The errors—found in an investigation published Saturday by the Independent, a progressive Canadian news outlet covering the country’s easternmost province Newfoundland and Labrador—appear in a 526-page report that was disseminated by its government in May. 

The report advised the then Liberal-led government’s Department of Health and Community Services on topics including virtual care, retention incentives, and the impacts of the COVID-19 pandemic on health care workers during a time when the province’s health care sector is facing nurse and doctor staffing shortages. 

The Deloitte report contained false citations, pulled from made-up academic papers to draw conclusions for cost-effectiveness analyses, and cited real researchers on papers they hadn’t worked on, the Independent found. It included fictional papers coauthored by researchers who said they had never worked together. 

“Deloitte Canada firmly stands behind the recommendations put forward in our report,” a Deloitte Canada spokesperson told Fortune in a statement. “We are revising the report to make a small number of citation corrections, which do not impact the report findings. AI was not used to write the report; it was selectively used to support a small number of research citations.”

The lengthy report also cited an academic paper from the Canadian Journal of Respiratory Therapy, which is yet to be found when searching its database.

“It sounds like if you’re coming up with things like this, they may be pretty heavily using AI to generate work,” Gail Tomblin Murphy, an adjunct professor in the School of Nursing at Dalhousie University in Nova Scotia, told the Independent. Tomblin Murphy was cited by Deloitte in an academic paper that “does not exist.” She added that she had worked with only three of the six other authors named in the false citation.

“And I definitely think that there’s many challenges with that. We have to be very careful to make sure that the evidence that’s informing reports [is] the best evidence, that it’s validated evidence. And that, at the end of the day, these reports—not just because they cost governments and they cost the public—[are] accurate and evidence-informed and helpful to move things forward.”

As of Monday, the report remains on the Canadian government’s website.

The Canadian government spent just under $1.6 million on the report, paying in eight installments, according to an access to information request published in a blog post last Wednesday. 

Tony Wakeham, leader of the Progressive Conservative Party in the province and the province’s new premier, was sworn into office in late October. Newfoundland and Labrador’s Office of the Premier and the province’s Department of Health and Community Services did not immediately respond to Fortune’s request for comment on the May report and has not publicly addressed the issue.

The revelation comes on the heels of news last month that Deloitte leveraged AI in a $290,000 report published in July to help the Australian government crack down on welfare. But a researcher flagged hallucinations in the 237-page study, which included references to nonexistent academic research papers and a fabricated quote from a federal court judgment.

In the revised study, which was quietly uploaded to the Australian government’s website, the consulting firm admitted it had used the generative AI language system Azure OpenAI to help create the report. 

“The updates made in no way impact or affect the substantive content, findings, and recommendations in the report,” Deloitte wrote in a section in the updated study.

Deloitte’s member firm in Australia was required to pay the government a partial refund for the report. No information has been made public yet in regards to a potential refund for Canada’s report.



Source link

Continue Reading

Business

Nearly three-quarters of Trump voters think the cost of living is bad or the worst ever

Published

on



President Donald Trump and his administration insist that costs are coming down, but voters are skeptical, including those who put him back in the White House.

Despite Republicans getting hammered on affordability in off-year elections last month, Trump continues to downplay the issue, contrasting with his message while campaigning last year.

“The word affordability is a con job by the Democrats,” Trump said during a Cabinet meeting on Tuesday. “The word affordability is a Democrat scam.”

But a new Politico poll found that 37% of Americans who voted for him in 2024 believe the cost of living is the worst they can ever remember, and 34% say it’s bad but can think of other times when it was worse.

The White House has said Trump inherited an inflationary economy from President Joe Biden and point to certain essentials that have come down since Trump began his second term, such as gasoline prices.

The poll shows that 57% of Trump voters say Biden still bears full or almost full responsibility for today’s economy. But 25% blame Trump completely or almost completely.

That’s as the annual rate of consumer inflation has steadily picked up since Trump launched his global trade war in April, and grocery prices have gained 1.4% between January and September.

Meanwhile, Vice President JD Vance pleaded for “patience” on the economy last month as Americans want to see prices decline, not just grow at a slower pace.

Even a marginal erosion in Trump’s electoral coalition could tip the scales in next year’s midterm elections, when the president will not be on the ballot to draw supporters.

A soft spot could be Republicans who don’t identify as “MAGA.” Among those particular voters, 29% said Trump has had a chance to change things in the economy but hasn’t taken it versus 11% of MAGA voters who said that.

Across all voters, 45% named groceries as the most challenging things to afford, followed by housing (38%) and health care (34%), according to the Politico poll.

The poll comes as wealthier households are having trouble affording basics, while discount retailers like Walmart and even Dollar Tree are seeing more higher-income customers.

And in a viral Substack post last month, Michael Green, chief strategist and portfolio manager for Simplify Asset Management, argued that the real poverty line should be around $140,000.

“If the crisis threshold—the floor below which families cannot function—is honestly updated to current spending patterns, it lands at $140,000,” he wrote. “What does that tell you about the $31,200 line we still use? It tells you we are measuring starvation.”



Source link

Continue Reading

Business

Apple is experiencing its biggest leadership shakeup since Steve Jobs died, with over half a dozen key executives headed for the exits

Published

on



Apple is currently undergoing the most extensive executive overhaul in recent history, with a wave of senior leadership departures that marks the company’s most significant management realignment since its visionary co-founder and CEO Steve Jobs died in 2011. The leadership exodus spans critical divisions from artificial intelligence to design, legal affairs, environmental policy, and operations, which will have major repercussions for Apple’s direction for the foreseeable future.

On Thursday, Apple announced Lisa Jackson, its VP of environment, policy, and social initiatives, as well as Kate Adams, the company’s general counsel, will both retire in 2026. Adams has been Apple’s chief legal officer since 2017, and Jackson joined Apple in 2013. Adams will step down late next year, while Jackson will leave next month.

Jackson and Adams join a growing list of top executives who have either left or announced their exits this year. AI chief John Giannandrea announced his retirement earlier this month, and its design lead Alan Dye, who took charge of Apple’s all-important user interface design after Jony Ive left the company in 2019, was just poached by Mark Zuckerberg’s Meta this week.​

The scope of the turnover is unprecedented in the Tim Cook era. In July, Jeff Williams, Apple’s COO who was long thought to succeed Cook as CEO, decided to retire after 27 years with the company. One month later, Apple’s CFO Luca Maestri also decided to step back from his role. And the design division, which just lost Dye, also lost Billy Sorrentino, a senior design director, who left for Meta with Dye. Things have been particularly turbulent for Apple’s AI team, though: Ruoming Pang, who headed its AI Foundation Models Team, left for Meta in July and took about 100 engineers with him. Ke Yang, who led AI-driven web search for Siri, and Jian Zhang, Apple’s AI robotics lead, also both left for Meta.

Succession talks heat up

While all of these departures are a big deal for Apple, the timing may not be a coincidence. Both Bloomberg and the Financial Times have reported on Apple ramping up its succession plan efforts in preparation for Cook, who has led the company since 2011, to retire in 2026. Cook turned 65 in November and has grown Apple’s market cap from about $350 billion to a whopping $4 trillion under his tenure. Bloomberg reports John Ternus has emerged as the leading internal candidate to replace him.​

Apple choosing Ternus would be a pretty major departure from what’s worked for Apple during the past decade, which has been letting someone with an operational background and a strong grasp of the global supply chain lead the company. Ternus, meanwhile, is focused on hardware development, specifically for the iPhone, iPad, Mac, and Apple Watch. But it’s that technical expertise that’s made him an attractive candidate, especially as much of the recent criticism about Apple has revolved around the company entering new product categories (Vision Pro, but also the ill-fated Apple Car), as well as its struggling AI efforts.​

Now, of course, with so many executives leaving Apple, succession plans extend beyond the CEO role. Apple this week announced it’s bringing in Jennifer Newstead, who currently works as Meta’s chief legal officer, to replace Adams as the company’s general counsel starting March 1, 2026. Newstead is expected to handle both legal and government affairs, which is essentially a consolidation of responsibilities among Apple’s leadership team, merging Adams’ and Jacksons’ roles into one.​

Alan Dye, meanwhile, will be replaced by Stephen Lemay, a move that’s reportedly being celebrated within Apple and its design team in particular. John Gruber, who’s reported on Apple for decades and has deep ties within the company, wrote a pretty scathing critique about Dye, but in that same breath said employees are borderline “giddy” about Lemay—who has worked on every major Apple interface design since 1999, including the very first iPhone—taking over.

Meanwhile, on the AI team, John Giannandrea will be replaced by Amar Subramanya, who led AI strategy and development efforts at Google for about 16 years before a brief stint at Microsoft.

Hitting the reset button

All of the above departures cover critical functions for Apple: AI competitiveness, design innovation, regulatory navigation, and operational efficiency. Each replacement brings specialized expertise that aligns with the challenges Cook’s successor will inherit.

The real test will be execution across multiple fronts simultaneously. Can Subramanya accelerate Apple’s AI development to match competitive threats? Will Lemay’s design leadership maintain Apple’s interface advantages as AI reshapes user interaction? Can Newstead navigate regulatory challenges while preserving Apple’s privacy-first approach?

What’s certain is the company will look fundamentally different in 2026—and the executive team that grew Apple into a $4 trillion behemoth is departing. The transformation could be as profound as any since Jobs handed the reins to his COO at the time, Tim Cook, 14 years ago.



Source link

Continue Reading

Business

Elon Musk says Tesla owners will soon be able to text while driving

Published

on



Elon Musk has given the thumbs up to some Tesla drivers texting behind the wheel.

The EV maker recently introduced a 30-day free trial of its Full Self-Driving (Supervised) (FSD) features on its North American cars, which has traffic-aware cruise control, autosteer, and autopark. To the Tesla CEO, the automated features in place are enough to condone texting while driving. According to safety experts, Musk’s suggestion is actually plain illegal.

In response to an X user’s question on Thursday about being able to text and drive while a Tesla is operating FSD v14.2.1, its latest full self-driving capabilities, Musk responded: “Depending on context of surrounding traffic, yes.”

Musk’s response mirrors his comments at Tesla’s annual shareholder meeting last month, where he said the company would soon feel comfortable with a multitasking driver.

“We’re actually getting to the point where we almost feel comfortable allowing people to text and drive, which is kind of the killer [application] because that’s really what people want to do,” Musk said. “Actually right now, the car is a little strict about keeping eyes on the road, but I’m confident that in the next month or two—we’re going to look closely at the safety statistics—but we will allow you to text and drive essentially.”

With a $1 trillion pay package on the line, Musk has worked to jumpstart Tesla after continued lagging sales. His lofty automation goals tied to the compensation plan include delivering 20 million vehicles and having 10 million active FSD subscriptions, as well as 1 million robotaxis on the commercially operational.

FSD roadbumps 

Tesla’s FSD rollout, much like its other automated technologies, has hit snags. In October, the U.S. Department of Transportation-run National Highway Traffic Safety Administration (NHTSA) opened an investigation into the EV maker, alleging its FSD software violated traffic laws and led to six crashes, four of which resulted in injuries. It cited data from 18 complaints from Tesla users claiming the FSD-equipped cars ran red lights or swerved into other lanes, including into oncoming traffic.

There is another complication for Musk’s vision of a Tesla owner typing away behind the wheel: Texting and driving is illegal in nearly the entire country, barring Montana, according to the U.S. Bureau of Transportation Statistics. According to the NHTSA, distracted driving resulted in 3,275 deaths in 2023.

Even Tesla has warned owners against texting while driving, even with some automated features in place: Tesla’s Model Y Owner’s Manual asks drivers not to use their phones while driving with Autopilot software enabled. (Autopilot refers to Tesla’s basic driver assistance features requiring hands on the steering wheel, while FSD is a paid subscription package with enhanced automated features and does not require a driver to have hands on the steering wheel.)

“Do not use handheld devices while using Autopilot features,” the manual said. “If the cabin camera detects a handheld device while Autopilot is engaged, the touchscreen displays a message reminding you to pay attention.”

Tesla did not respond to Fortune’s request for comment.

What experts are saying

Alexandra Mueller, senior research scientist for Insurance Institute for Highway Safety, told Fortune condoning texting while behind the wheel completely undermines the purpose of Tesla’s current automated features Tesla, which are a level 2 on the five-point automation scale, meaning the models require the driver to still be fully in control of the vehicle.

“Having partial automation support doesn’t mean that you suddenly can kick back and text and not worry about driving,” Mueller said, “because that’s just not how these systems are designed to be used—and that’s also not the responsibility that the driver has when using these systems, and that’s by design.”

She said automated systems like Tesla’s are not designed to replace the driver and work because they are “human-in-the-loop” and were designed to support the driver’s discretion behind the wheel. Beeps and notifications from the vehicle if a driver changes lanes without signalling can help shape good behaviors, Mueller noted. Encouraging multitasking behind the wheel turns these features into convenience factors, rather than the safety precautions they were intended to be.

“Suddenly all your safety assessments on the technology don’t apply anymore, because you’ve changed the very nature of how the technology is supporting human-in-the-loop behavior,” Mueller concluded.



Source link

Continue Reading

Trending

Copyright © Miami Select.