“A human mathematician has always undergone an extensive training. This training may be regarded as not unlike putting instruction tables into a machine,” Turing said in a lecture to the London Mathematical Society. “One must therefore not expect a machine to do a very great deal of building up of instruction tables on its own.”

Proposing the game of chess, Turing added: “The machine must be allowed to have contact with human beings in order that it may adapt itself to their standards.” 

Hmm. Humanity’s standards? Can’t decide if that’s hopeful or bleak, Prof.

Today’s news below. —Andrew Nusca

Want to send thoughts or suggestions to Data Sheet? Drop a line here.

Microsoft unveils Majorana 1 quantum processor

Microsoft is doing its part to bring quantum computing closer to reality.

The tech giant on Wednesday unveiled a new chip called Majorana 1 that it says is not just powerful—duh—but reliable enough to actually make quantum computing usable in the real world.

What might one use quantum computing for? Drug discovery, weather forecasting, financial risk analysis, logistics optimization, and any other problem that is simply too complex and/or inefficient for today’s traditional computing hardware—even supercomputers.

Microsoft says Majorana 1 uses a different approach to quantum, called “topological,” that makes the chips more scalable than what’s on offer from category leaders IBM or Google. (Though those companies, not to mention some researchers, may object to that.) 

Its intermediate results were published in the scientific journal Nature.

Microsoft CEO Satya Nadella said the advancement promises “meaningful” quantum computers in years rather than decades—good news for the U.S. and China, who are battling for technological supremacy in the category through both investments and trade restrictions.

“Sometimes researchers have to work on things for decades to make progress possible,” Nadella wrote in a social media post. “It takes patience and persistence to have big impact in the world.” —AN

Apple introduces iPhone 16e

Apple has refreshed the low end of its smartphone lineup with the iPhone 16e.

The from-$599 handset is notable for a bunch of reasons. First off, it’s not called an iPhone SE, like Apple’s previous “cheap” iPhones—and with a 6.1” display, it should provide a much nicer viewing experience.

The iPhone 16e packs enough grunt to handle Apple Intelligence, as all the company’s devices must these days—there’s an A18 chip in there, just like with the regular, current iPhone 16. This will be important for ensuring that Apple’s AI can roll out to more of the market.

The 16e even has Apple’s recently-introduced Action button, which users can program to activate a variety of apps and functions. (Apple’s home button, last seen in the 2022 iPhone SE, has now officially shuffled off this mortal coil.)

But the 16e is also important because it’s the first iPhone to use Apple’s homegrown cellular modem, the C1. The regular iPhone 16 uses Qualcomm’s Snapdragon X75 5G modem.

So why get the from-$799 iPhone 16 instead of the much cheaper 16e? The 16e only comes with one camera (albeit one with “integrated 2x telephoto”) rather than two, and it also lacks its slightly older sibling’s Dynamic Island.

Meanwhile, in a striking coincidence, detailed specs of Google’s upcoming Pixel 9a—which will target the same segment of the market as the iPhone 16e—leaked just before Apple’s announcement. —David Meyer

Rivian’s $6.6 billion federal loan now hangs in the balance

Rivian’s most ambitious investment project to date could be at risk, after Georgia Gov. Brian Kemp warned disbursement of a $6.6 billion U.S. federal loan had been put on hold.

The automaker secured the financing through a dedicated Department of Energy program in the dying days of the Biden administration. The funds are needed to erect a manufacturing plant that can build its next-generation R2 electric cars at scale.

To date, the most famous recipient of the DOE program’s funding is Tesla. Even though the conditions attached can be onerous—CEO Elon Musk chose to repay his as early as possible—that didn’t stop the Trump transition team from pledging to investigate the deal.

Rivian is set to publish key fourth-quarter figures after the close of markets on Thursday. CEO and founder RJ Scaringe has promised to deliver a quarterly gross profit that could bolster confidence his business is truly viable.

Until now, the more EVs Rivian sells, the bigger the financial loss it reports, as the cost of its goods exceeds their average selling price. 

During the third quarter, Rivian posted a gross loss of $39,000 on each vehicle sold before headwinds were factored in like operating expenses and depreciation. Below the line, it was roughly $1.1 billion in the red for the period, all things considered.

Scaringe needs to fix this before he begins manufacturing 400,000 units of his R2 and smaller R3 crossover at the Georgia plant—assuming the loan is still disbursed and construction is completed. —Christiaan Hetzner

More data

—NetEase directed to divest. China’s second-largest game publisher unloads pricey U.S. developers.

—Microsoft debuts Muse AI, a generative AI model for game environments.

—Gemini stands alone. Google’s chatbot moves exclusively to a standalone app on iOS.

—US Army soldier pleads guilty to AT&T, Verizon hack. Data from 100 million customers in 2024.

—Meta’s Instagram adds real-time translation to DMs, with support for 99 languages.

—Perplexity’s R1 1776 AI model is an open-source DeepSeek variant “post-trained to provide unbiased, accurate, and factual information.” 

—Alibaba approached DeepSeek about an investment, but the company is being cautious.

—NIST prepares for widespread cuts. The agency promotes U.S. innovation and competitiveness.

—Sanas uses AI to change the accents of call center employees in real time.

—Libra cofounder bragged about controlling Argentina’s president via payments to sister. An anti-corruption investigation is underway.

Endstop triggered