What we’re reading (6/9)
“A Bond Market Meltdown Might Be Inevitable” (The Hill). “Over the past five years, both the budget and trade deficits have deteriorated sharply. Budget deficits have exceeded 5 percent of GDP since 2020 and projections indicate deficits will remain elevated, raising concerns about fiscal sustainability. Critically, government borrowing costs have risen sharply since 2022. Historian Niall Ferguson has suggested that America’s superpower status may be threatened as the U.S. government now spends more on interest payments than on defense. “
“The Canned-Food Aisle Is Getting Squeezed By Rising Steel Tariffs” (Wall Street Journal). “The Trump administration’s new 50% duty on imported steel could increase store prices for items in steel cans by 9% to 15%, according to the Consumer Brands Association, a trade group whose members include Campbell’s, Hormel Foods and Del Monte Foods. At that rate, the price of a can of vegetables costing $2 could increase by 18 cents to 30 cents. ‘The American consumer is going to pay more for their cans,’ said Dan Dietrich, vice president for strategy at Trivium Packaging.”
“We’re Secretly Winning The War On Cancer” (Vox). “The age-adjusted death rate in the US for cancer has declined by about a third since 1991, meaning people of a given age have about a third lower risk of dying from cancer than people ofthe same age more than three decades ago. That adds up to over 4 million fewer cancer deaths over thattime period. Thanks to breakthroughs in treatments like autologous stem-cell harvesting and CAR-T therapy…cancer isn’t the death sentence it once was.”
“The Illusion Of Thinking: Understanding The Strengths And Limitations Of Reasoning Models Via The Lens Of Problem Complexity” (Apple Machine Learning Research). “By comparing LRMs with their standard LLM counterparts under equivalent inference compute, we identify three performance regimes: (1) low- complexity tasks where standard models surprisingly outperform LRMs, (2) medium-complexity tasks where additional thinking in LRMs demonstrates advantage, and (3) high-complexity tasks where both models experience complete collapse. We found that LRMs have limitations in exact computation: they fail to use explicit algorithms and reason inconsistently across puzzles. We also investigate the reasoning traces in more depth, studying the patterns of explored solutions and analyzing the models’ computational behavior, shedding light on their strengths, limitations, and ultimately raising crucial questions about their true reasoning capabilities.”
“Why Apple’s Stock Suddenly Dropped A Few Minutes Into Its WWDC Keynote” (Business Insider). “About six minutes into the presentation, Apple's stock abruptly fell more than 2.5%, from roughly $206 to below $201. That's the equivalent of about $75 billion in market value. Just seconds earlier, Apple software chief Craig Federighi was on stage recapping the Apple Intelligence features the company had already rolled out in recent months, such as Genmoji, smart replies, photo cleanup tools. Then he pivoted to Siri. That's when things got awkward.”
