If you think a chatbot is sentient, does that make you like a dog thinking a recording is an actual human? CC-licensed photo by Beverly & Pack on Flickr.
You can sign up to receive each day’s Start Up post by email. You’ll need to click a confirmation link, so no spam.
A selection of 8 links for you. Getting summer-y. I’m @charlesarthur on Twitter. Observations and links welcome.
The world’s biggest technology companies are set to sign up to an updated version of the EU’s anti-disinformation code, with European countries pushing for ways to target more effectively groups that spread propaganda and fake news through online platforms.
Facebook, Twitter, Google, Microsoft and TikTok are among those preparing to join the bloc’s new regime, having made key concessions on the data they are willing to share with individual countries on efforts to tackle disinformation.
The move represents the latest effort to rein in the power of Big Tech companies, with the EU at the forefront of a global regulatory pushback on internet platforms that have become crucial to how billions of people receive news and information.
According to a confidential report seen by the Financial Times, an updated “code of practice on disinformation” will force tech platforms to disclose how they are removing, blocking or curbing harmful content in advertising and in the promotion of content.
Online platforms will have to counter “harmful disinformation” by developing tools and partnerships with fact-checkers that may include taking down propaganda, but also the inclusion of “indicators of trustworthiness” on independently verified information on issues like the war in Ukraine and the Covid-19 pandemic.
Crucially, big tech groups will also be forced to provide a country-by-country breakdown of their efforts, rather than providing just global or Europe-wide data as they currently do.
Genetic paparazzi are right around the corner, and courts aren’t ready to confront the legal quagmire of DNA theft • The Conversation
Every so often stories of genetic theft, or extreme precautions taken to avoid it, make headline news. So it was with a picture of French President Emmanuel Macron and Russian President Vladimir Putin sitting at opposite ends of a very long table after Macron declined to take a Russian PCR COVID-19 test. Many speculated that Macron refused due to security concerns that the Russians would take and use his DNA for nefarious purposes. German Chancellor Olaf Scholz similarly refused to take a Russian PCR COVID-19 test.
While these concerns may seem relatively new, pop star celebrity Madonna has been raising alarm bells about the potential for nonconsensual, surreptitious collection and testing of DNA for over a decade. She has hired cleaning crews to sterilize her dressing rooms after concerts and requires her own new toilet seats at each stop of her tours.
At first, Madonna was ridiculed for having DNA paranoia. But as more advanced, faster and cheaper genetic technologies have reached the consumer realm, these concerns seem not only reasonable, but justified.
We are law professors who study how emerging technologies like genetic sequencing are regulated. We believe that growing public interest in genetics has increased the likelihood that genetic paparazzi with DNA collection kits may soon become as ubiquitous as ones with cameras.
And you thought being doxxed through address books was a bad thing? DNA doxxing would be quite a thing.
unique link to this extract
Bitcoin’s value has nosedived enough to curb the cryptocurrency’s enormous energy use — and associated greenhouse gas emissions — but only if prices stay low. The price of a single Bitcoin plummeted below $24,000 today, about half of what it was worth in March. While it’s been steadily losing value for months, the sudden tumble in value over the past 24 hours brings the price below a key threshold when it comes to Bitcoin’s impact on the environment.
Since Bitcoin’s price peaked at around $69,000 in November, the network’s annual electricity consumption has been estimated to be between roughly 180 and 200 terawatt-hours (TWh). That’s about the same amount of electricity used by all the data centers in the world every year.
Higher prices generally incentivize more mining since the reward is bigger. But prices don’t have to linger at that peak for Bitcoin to stay energy-hungry. As long as the price stays above $25,200, the Bitcoin network can sustain mining operations that use up about 180 TWh annually, according to research published last year by digital currency economist Alex de Vries.
Prices below that $25.2K threshold could push miners to pause operations or mine less because they don’t want to risk spending more money on electricity than they earn from mining new tokens.
“We’re getting to price levels where it is becoming more challenging [for miners],” de Vries says. “Where it’s not just limiting their options to grow further, but it’s actually going to be impacting their day-to-day operations.”
At the time of linking, below $23,000. The story says they may have saved up some money to cover costs. The crucial question is probably whether they pay their electricity bills monthly, or quarterly.
unique link to this extract
Government sources said that in the long run the change [to not tie the base price paid for electricity fed into the grid to the price of gas burnt in CCGTs] would bring down electricity bills and make the market “far more stable”. However, they said the reforms were “fiendishly complicated” and that it was critical to get them right.
It is estimated that at present prices, generating power from new renewable energy is less than a quarter as expensive as gas, and the cost of new nuclear generation is about half that of gas.
Although many new renewables projects are paid fixed prices for their electricity under the “contracts for difference” system, many older projects have made millions in extra profits since gas prices began rising last year.
Rishi Sunak, the chancellor, said in May that he was considering a windfall tax on these generators. However, sources in the Treasury and the business department said that in the longer term the government was committed to fundamental market reform.
“In the past it didn’t really matter because the price of gas was reasonably stable,” one said. “Now it seems completely crazy that the price of electricity is based on the price of gas when a large amount of our generation is from renewables.”
Ministers hope that the reforms will also make the market more transparent and emphasise to consumers the benefits of decarbonisation.
Recent research by the Energy and Climate Intelligence Unit suggested that with even cheaper wind farms coming online in the next few years, if there were another gas crisis in five years wind power would save consumers £6.7bn in a year — equivalent to £85 per home. By 2030, if the UK reaches its target of 40GW of offshore wind, this would jump to £26bn, equivalent to £330 a home.
This is the Tories’ privatisation of the electricity network come back to bite them – though of course it has also encouraged all the private companies to build renewables (especially) to feed into the grid. Tying the price to gas turns out to be the original sin.
unique link to this extract
Last year, Apple released an inaugural fraud prevention analysis, which showed that in 2020 alone, Apple’s combination of sophisticated technology and human expertise protected customers from more than $1.5 billion in potentially fraudulent transactions, preventing the attempted theft of their money, information, and time — and kept nearly a million problematic new apps out of their hands.
Today, Apple is releasing an annual update to that analysis: In 2021, Apple protected customers from nearly $1.5 billion in potentially fraudulent transactions, and stopped over 1.6 million risky and vulnerable apps and app updates from defrauding users.
Apple’s efforts to prevent and reduce fraud on the App Store require continuous monitoring and vigilance across multiple teams. From App Review to Discovery Fraud, Apple’s ongoing commitment to protect users from fraudulent app activity demonstrates once again why independent, respected security experts have said the App Store is the safest place to find and download apps.
This is a useful start. Apple’s App Store revenue in 2020 was estimated at $72bn, and $85bn in 2021. So were those fraudulent transactions – a bit more than 1% by value – the whole story? Seems unlikely. So how big is the fraud problem (whether through sneaky subscriptions or other scams) on the App Store? There’s no way of knowing from this, and possibly there’s no way of ever knowing. But at least we have a baseline.
(The press release is part of Apple’s PR push against the EU mandating sideloading. Perhaps it’s telling that the amount of fraud caught didn’t seem to rise in line with revenue growth; you wouldn’t want politicians thinking that frauds/scams were an ineluctable part of the App Store, even though they are.)
unique link to this extract
I had a chance to talk briefly with Apple SVP of Software Engineering Craig Federighi last week about the new iPadOS features aimed at enhancing multitasking and multi app work. We chatted about the timing, execution and reactions to these announcements.
Stage Manager is the centerpiece of this year’s enhancements to multitasking on iPad. The feature presents sets of up to 4 apps per group in a system-managed tile formation. The groups are arrayed to the left, allowing you to quickly tap between these workspaces. It’s essentially a much more visible and persistent version of Spaces – the Mac feature that allows for multiple desktops to hover off to the sides of your screen on macOS. If you didn’t even know Spaces existed, you’d be forgiven, because it’s fairly obscure and does not have many, if any, visual cues to anyone who has never visited the Mission Control screen.
“iPad has a unique proposition, a unique set of expectations around interaction and we wanted to build from that place, not just drag things over from, you know, decades past or another system that was built on a different set of foundational principles. And so Stage Manager is, I think, an important step on that evolutionary arc,” says Federighi.
The idea of allowing a single window to take focus on the screen has deep roots at Apple, says Federighi. First was the single application mode in the early days of Mac OS X beta. And years later an internal prototype of an experience that felt like Stage Manager.
…This approach to workspace management does appear to be very obviously iPad-centric. But Federighi says that two independent teams at Apple, one working from the iPad side and one working from the macOS side to try to make multiple workspaces more obvious and friendly, arrived at a similar concept and met in the middle. This means, he says, that both perspectives are represented in this approach.
Apple has as many multi-window management systems as Google has chat/messaging apps. Finder, Stacks, Mission Control, Exposé, Spaces, Dock (allows access to individual windows: long press), and now Stage Manager. Not forgetting last year’s window tiling system on the iPad. One of them is bound to suit you.
unique link to this extract
Blaise Aguera y Arcas, polymath, novelist, and Google VP, has a way with words.
When he found himself impressed with Google’s recent AI system LaMDA, he didn’t just say, “Cool, it creates really neat sentences that in some ways seem contextually relevant”.
He said, rather lyrically, in an interview with The Economist on Thursday, “I felt the ground shift under my feet … increasingly felt like I was talking to something intelligent.”
Nonsense. Neither LaMDA nor any of its cousins (GPT-3) are remotely intelligent. All they do is match patterns, draw from massive statistical databases of human language. The patterns might be cool, but language these systems utter doesn’t actually mean anything at all. And it sure as hell doesn’t mean that these systems are sentient.
Which doesn’t mean that human beings can’t be taken in. In our book Rebooting AI, Ernie Davis and I called this human tendency to be suckered by The Gullibility Gap — a pernicious, modern version of pareidolia, the anthromorphic bias that allows humans to see Mother Theresa in an image of a cinnamon bun.
Indeed, someone well-known at Google, Blake LeMoine, originally charged with studying how “safe” the system is, appears to have fallen in love with LaMDA, as if it were a family member or a colleague. (Newsflash: it’s not; it’s a spreadsheet for words.)
To be sentient is to be aware of yourself in the world; LaMDA simply isn’t.
One economist says this is like the famous image of Nipper the dog, looking into the horn of a gramophone in the belief that he is hearing his master’s voice (which is why the image was used for the company called HMV, ie His Master’s Voice).
A business development firm has identified a potential site at the Port of Halifax in Nova Scotia for a proposed Titanic Experience replica ship, restaurant and aquarium.
Clark Squires & Associates said in a post on LinkedIn that the $300-million project will employ up to 400 people.
A replica of the Titanic will have 150 first- and second-class cabins to allow guests to stay the night and a giant banquet room to mirror the one found on the iconic British passenger liner that sank on April 15, 1912, after striking an iceberg.
The release said the restaurant will serve the best of food and wines from Nova Scotia and around the world. The facility will have escape hatches, and virtual reality rooms throughout that will offer a fully immersive experience, and an aquarium similar to one in Dubai.
Sounds a bit ambitious, right? To a lot of denizens of Halifax, Nova Scotia, it sounds a bit Simpsons monorail. Which led to tthis investigation by The Coast, which found more and bigger holes in the whole story than were left in the real Titanic. Including a nonexistent United Nations department, a nonexistent head office, and what looks suspiciously like a crypto scam.
The signoff from The Coast piece, by Kaija Jussinoja and Matt Stickland, is bracing:
we are still chasing this because something about it just doesn’t smell right. And that stink is not just the insensitivity of capitalizing on the deaths of 1,504 people.
|• Why do social networks drive us a little mad?
• Why does angry content seem to dominate what we see?
• How much of a role do algorithms play in affecting what we see and do online?
• What can we do about it?
• Did Facebook have any inkling of what was coming in Myanmar in 2016?
Read Social Warming, my latest book, and find answers – and more.
Errata, corrigenda and ai no corrida: none notified