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.

»

Securus Technologies, a prison telecommunications company previously reported on for its predatory business practices, deleted incarcerated users’ draft emails in a system reboot on Monday. One incarcerated journalist says they use the draft box to do critical reporting and lost years of work as a result of the wipe, and that the company silenced important reporting. The company says it was an accident.

Christopher Blackwell, an incarcerated journalist at the Washington Corrections Center in Washington State, posted on X (formerly Twitter) on Monday that Securus had wiped his draft email box. Blackwell has written for numerous outlets about the conditions he faces while incarcerated—he’s active on X by dictating posts over the phone. 

“I’m an incarcerated journalist,” Blackwell wrote in the post. “Today Securus, a predatory prison communication co, silenced journalists. With no warning they deleted all drafts of writing. Years of work. Manuscripts. Articles. Everything gone. We’re no longer able to save drafts. Now near impossible to write.” 

Securus provides inmates with tablets that allow them to access their emails, as well as direct messaging and video-chats on the platform. Blackwell told Motherboard in a phone call that Securus charges inmates 24 cents per email, either coming in or going out. He said it was a “common occurrence” for Securus to reboot its system, and that this was the fourth time it had happened this year. This time, however, Securus also changed its settings so that no draft email could be saved for longer than 24 hours, which he said had never been the case previously. 

“For anybody who’s an inside journalist, this is a really big deal,” Blackwell said. “I do a lot of investigative pieces—5,000 or 6,000 words. If I can’t save anything, how can I do anything? It’s basically a full-out attack for them to block us from doing these kinds of pieces, and it’s an under-handed way of doing it. This is just one of the many things that they do.”

Both the Washington Department of Corrections and Securus told Motherboard in statements that the draft box deletion and saving settings change had been a mistake. As compensation, the company gave inmates two “stamps” each—an internal system currency for making communications. 

«

»

Giraffe populations have declined by 40% in the last 30 years, and there are now fewer than 70,000 mature individuals left in the wild. What are the causes of this alarming decline, and what can be done to protect these gentle giants?

The five biggest threats to giraffes are habitat loss, insufficient law enforcement, ecological changes, climate change, and lack of awareness. Below, I will tell you about these threats and what is being done to save them.

I will also explain a study I was a part of that ranked these threats in terms of each one’s danger of causing giraffe extinction, and whether human actions can alleviate that danger. The study used data from more than 3,100 giraffes identified over eight years in an unfenced 4,500km² area of the Tarangire ecosystem in Tanzania. We used the data to simulate how environmental and land use changes could affect the giraffe population over 50 years.

…Given their vast historical Africa-wide range and individual home ranges of thousands of hectares, giraffes will not likely survive only within the boundaries of small, fragmented protected areas. I propose as part of our evidence-based recommendations that rangelands used by wildlife and pastoralists as movement pathways be permanently protected from farming, mining and infrastructure. This will give people as well as wide-ranging animals like giraffes freedom to roam.

It will also require the expansion of wildlife law enforcement in village lands outside formal protected areas.

«

»

The global AI safety summit opened at Bletchley Park on Wednesday with a landmark declaration from countries including the UK, US, EU and China that the technology poses a potentially catastrophic risk to humanity.

The so-called Bletchley declaration said: “There is potential for serious, even catastrophic, harm, either deliberate or unintentional, stemming from the most significant capabilities of these AI models.”

«

»

Inhaling polluted air increases the risk of type 2 diabetes, the first study of its kind in India has found. Research conducted in Delhi and the southern city of Chennai found that inhaling air with high amounts of PM2.5 particles led to high blood sugar levels and increased type 2 diabetes incidence.

When inhaled, PM2.5 particles – which are 30 times thinner than a strand of hair – can enter the bloodstream and cause several respiratory and cardiovascular diseases.

The study is part of ongoing research into chronic diseases in India that began in 2010. It is the first to focus on the link between exposure to ambient PM2.5 and type 2 diabetes in India, one of the worst countries in the world for air pollution.

The average annual PM2.5 levels in Delhi was 82-100μg/m3 and in Chennai was 30-40μg/m3, according to the study, many times the WHO limits of 5μg/m3. India’s national air quality standards are 40μg/m3.

There is also a high burden of non-communicable diseases, including diabetes, hypertension and heart disease in India; 11.4% of the population – 101 million people – are living with diabetes, and about 136 million are pre-diabetic, according to a paper published in the Lancet in June.

«

»

The European Data Protection Board said Wednesday it had adopted a binding decision that will ban Facebook and Instagram owner Meta from using the personal data of users for targeted ads without their explicit consent.

The order closes off the legal basis used by Meta to freely process the personal data of its users in Europe to target ads, a practice that has made the company founded by Mark Zuckerberg one of the world’s most profitable companies.

Instead, Meta will now need clear consent by the user to keep harvesting their data, a new protocol which will also face intense legal scrutiny among privacy activists and European regulators.

The latest order will “impose a ban on the processing of personal data for behavioural advertising on the legal bases of contract and legitimate interest across the entire European Economic Area,” the EDPB said.

Firms like Meta use user data to serve highly targeted ads, and they have struggled to comply with the strict rules of the EU’s 2018 data privacy regulation (GDPR).

In anticipation of the order, Meta said on Monday that Facebook and Instagram users in Europe will be able to buy subscriptions to use the social networks without any advertising from this month.

Meta believes that taking this move to offer subscriptions will assuage EU regulators’ concerns over its data collection and how ads are targeted.

“Meta has already announced that we will give people in the EU and EEA the opportunity to consent and, in November, will offer a subscriptions model to comply with regulatory requirements,” a Meta spokesperson said.

«

»

Scarlett Johansson has taken legal action against an AI app that used her name and likeness in an AI-generated advertisement without her permission.

The 22-second ad, posted on X, the platform formerly known as Twitter, by an image generating app called Lisa AI: 90’s Yearbook & Avatar, used real footage of Johansson to generate a fake image and dialogue for her.

Representatives for the 38-year-old actor confirmed to Variety that she is not a spokesperson for the company and that appropriate legal actions were taken since it was spotted on 28 October. “We do not take these things lightly. Per our usual course of action in these circumstances, we will deal with it with all legal remedies that we will have,” her lawyer, Kevin Yorn told Variety. The ad has since been taken down.

The video, reviewed by Variety, opened with a behind the scenes clip of Johansson from the set of the Marvel film Black Widow. “What’s up guys? It’s Scarlett and I want you to come with me … ” she says, before the screen transitions into AI-generated photos that resemble her. A voice imitating the actor continues speaking to promote the app: “It’s not limited to avatars only. You can also create images with texts and even your AI videos. I think you shouldn’t miss it,” the fake narration says.

The fine print under the advertisement says: “Images produced by Lisa AI. It has nothing to do with this person.” Multiple apps by Lisa AI remain on the App Store and Google Play.

Johansson is not the only actor to speak out publicly against the use of their name and likenesses for AI. Last month, Tom Hanks took to Instagram to warn fans about a dental plan that used an AI-generated image of him for their promotional video. “Beware! … I have nothing to do with it,” he wrote on his Instagram story.

«

»

Online church services using artificial intelligence are rapidly becoming an essential part of worship in Korea, where Christianity is the biggest religion, as tens of thousands turn to chatbots and audio bibles for spiritual sustenance.

This year, local startups have launched generative AI bible study and prayer service apps, which in particular pull in young Protestants.

Pastors have welcomed the time the technology frees up for them to take care of their flock, who account for about a fifth of South Korea’s 52mn population. But they are also conscious of the need to maintain the human touch and have cautioned against relying too much on the latest technology for religious activities.

Awake Corp, the developer of ChatGPT-based bible chatbot service Ask Jesus — now rebranded as Meadow — has since its launch in March attracted about 50,000 users, including 10,000 from outside Korea. The app has drawn Christians in Muslim countries such as Pakistan as well as in the US and other western countries.

The service responds to inquiries on spiritual matters and day-to-day issues with bible verses, interpretations and prayers.

The app has generated interest from churches and pastors, who use Awake’s AI-driven WeBible web service to write sermons. When a pastor asks about a certain section of the Bible, the service can offer detailed explanations, identify main messages and points of reflection, and suggest a title for the sermon.

«

»

Tony Fadell, one of the creators of Apple’s defining product, the iPod, recently reminded us that the music player was launched on October 23, 2001. As I was in the process of moving back to New York, I bought one from a local Apple dealer. Soon, I observed an interesting social behavior upon my arrival: people with white headphones nodded at each other in public, as if acknowledging a shared coolness. We were all part of a tribe — smaller than most realized. It was weird kind of a snobbery. Today, while Apple seems ubiquitous, back then, it was distinctly the underdog.

I quickly penned a short, back-of-the-book piece titled “iPoddery” for Red Herring magazine. Over the following decades, I faithfully bought and upgraded every model of the iPod. I still have several lying around in my apartment. Yet, it was the iPod Shuffle that truly captured my heart.

It inspired me to develop a life philosophy I call “the iShuffle Principle”, long before Marie Kondo introduced her Live Better strategy. To summarize the principle for those who might skip the piece: “More is just more! Often, small and mostly predictable things are the most fun. That’s the iShuffle principle.” Adhering to this philosophy isn’t easy — consumerism is a curse. Living by the iShuffle principle requires constant effort and is an ongoing journey.

«

»

We talk a lot about how AI will transform products, industries, and everyday work and life, but what about particular business models?

• Increasing perversion of certain business models that are liable to be gamed or constrained by AI: …will Tinder or Bumble constrain the efficiency of AI so the product doesn’t become too “unsustainably effective”? Or in the world of music streaming: Since Spotify pays artists per song, will Spotify automatically optimize its algorithms to favor longer songs, taking into account the number of minutes each customer listens per day?

• Time-based business models are liable for disruption via a value-based overhaul of compensation: …most designers, lawyers, and many trades in between continue to charge by the hour, the AL-powered step-function improvements in workflows are liable to shake things up. Let’s first tackle this by considering the ultimate SOURCE of the differentiating value delivered to a client: It is less “time” and it is more “experience.”

«

»

While to start with, the app’s existence largely went under the radar, downloads of BeReal eventually skyrocketed, leaping from 1.1 million in February 2022 to an estimated 53 million by October in the same year. Its growth was largely driven by Gen Z, with under-25s making up almost 80% of users in some markets.

But analysis by various market intelligence firms this year has suggested that both monthly downloads and daily users are now in decline. According to analytics firm Apptopia, the number of people who use the app daily has dropped 61% from its peak, from about 15 million in October 2022 to less than six million in March 2023. The conversation around BeReal has become more jaded too. “BeReal going off at 10.30pm is it,” one X user wrote. “Suppose I better obediently take a picture of myself barely awake whilst laying down on the sofa watching TV alone so that the only three friends I have left who are using it can see.” But why have young people turned their backs on the app that was once the hot new thing?

For one, there’s a bossiness to the way in which BeReal operates. It demands that you post within two minutes of the notification, or else it snitches on you to your friends. It also locks your feed until you have posted your own BeReal, like a parent bribing a child. But young people don’t want to be parented by their social media. As Aisha Attah, 20, says, “I’m busy and like to use social media on my terms and in my own time.” Amid a wealth of other apps that you can engage with however you want, BeReal can feel too much like hard work. 

Dr Harry Dyer, digital sociologist and lecturer in education at the University of East Anglia, notes that by rushing people to post, BeReal may actually stress out some of its users, particularly Gen Z. “There’s a tonne of research on Gen Z hating phone calls” because they find “the idea of having to respond in that moment […] stressful”. BeReal, he says, may trigger a similar response by calling for that same kind of in-the-moment communication. He also believes that in “gatekeeping” the experience of browsing until users have themselves posted, the app disregards “a really important part of why we use social media” – the desire to ‘lurk’ and observe others while going unobserved ourselves.

«

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.

»

It’s agreed by both sides that Apple sought a meeting with Masimo when it was exploring adding pulse oximetry to the Apple Watch. Apple signed a confidentiality agreement. A meeting was held in 2013. At this point, stories differ. Joe Kiani, the electrical engineer who founded Masimo in 1989 and remains its chief executive officer, told me on Friday that Apple decided that rather than pay for licenses to use the technology, do a joint venture, or buy Masimo, Cercacor or both, it decided to do “strategic hiring” of key employees. Apple says it concluded that Masimo’s technology wasn’t suited for a consumer device, so it went in a different direction. (Although of course it did hire those key Masimo employees.)

Apple finally introduced its first watch with pulse oximetry, the Series 6, in 2020. It has incorporated pulse oximetry into all of its smart watches since then, with the exception of the cheaper Apple Watch SE. That means Apple will have to stop selling all its watches except the SE model starting on Christmas Day unless something happens before then: Either the companies cut a deal, the U.S. trade representative’s office recommends to Biden that he reverse the ruling, or Apple persuades the Federal Circuit Court of Appeals to stay the injunction while it considers an appeal.

The I.T.C. case is only one front in the battle between Masimo and Apple. In 2020, Masimo sued Apple in Federal District Court in California, alleging theft of trade secrets. The suit ended in a mistrial in May when the jury deadlocked. The case is being retried. Last year, Apple sued Masimo in federal court in Delaware, charging that Masimo was the one doing the illegal copying and was trying to knock Apple out of the market to make way for its own watch.

«

»

WeWork is planning to file for bankruptcy as early as next week, according to people familiar with the matter, in what would mark a stunning reversal for the flexible-office-space venture that was once valued at $47bn. 

New York-based WeWork is considering filing a chapter 11 petition in New Jersey, the people said. 

WeWork missed interest payments owed to its bondholders on Oct. 2, kicking off a 30-day grace period in which it needs to make the payments. Failing to do so would be considered an event of default. On Tuesday, the company said it has struck an agreement with the bondholders to allow it another seven days to negotiate with the stakeholders before a default is triggered. 

WeWork declined to comment on what a spokesperson called “speculation.” The spokesperson also pointed to a securities filing early Tuesday that “the forbearance agreement provides time to continue in the positive conversations with our key financial stakeholders and engage with them to implement our ongoing strategic efforts to enhance our capital structure.”

In August, the company shook up its board after three directors resigned due to a material disagreement regarding board governance and the company’s strategic direction, according to a securities filing. WeWork appointed four new directors with expertise in large, complex financial restructurings. Those directors have been negotiating with WeWork’s creditors over the past several months about a restructuring plan as they prepare for the bankruptcy.

«

»

Google is disabling live traffic conditions in Israel and Gaza on its Maps and Waze apps, a spokesperson said Tuesday. The move comes as Israel is widely anticipated to embark on a ground invasion into Gaza.

“As we have done previously in conflict situations and in response to the evolving situation in the region, we have temporarily disabled the ability to see live traffic conditions and busyness information out of consideration for the safety of local communities,” a Google spokesperson said in a statement. “Anyone navigating to a specific place will still get routes and ETAs that take current traffic conditions into account.”

The spokesperson said Google is working with local authorities as part of the ongoing Israel-Hamas War.

The Israel Defense Forces did not immediately respond to comment about whether it requested the restriction.

Google similarly disabled Maps live traffic data in Ukraine last year at the start of the Russia-Ukraine War, according to Reuters.

«

»

If you step an inch outside the weird and specific context of the relationship between Amazon and its millions of sellers, the company’s pitch for automated product photos sounds sort of insane. It’s a tool for faking product photos! It’s the sort of thing an e-commerce site might make a rule against using, and with good reason. Take the product photos in Amazon’s video, for example.

At first glance — which, for a thumbnail in an ad on Amazon, might be what matters most — things look good. We see a toaster in the middle of a sunlit kitchen, another on the edge of a stone table, another on some planks, and then on a counter again, except, hold on, this counter appears to extend all the way back to some drawers and might also be … the floor? I can see how such photos might result in better click-through rates for an ad, since they mimic certain popular aesthetics and imply a level of polish and care. What I can’t see is how big this toaster is. These photos somehow offer less information than the customary floating-in-white-space Amazon product shot. They don’t add context. They subtract it. This toaster has been banished to four different locations in the uncanny valley, none of which are anywhere near an electrical outlet.

Next, we see images created with the prompt “product in a kitchen, used in meal preparation.” Once more, regarding probably the main question that real-world product shots can help answer: how large is this appliance? On the top left, staged in a pizza-oven crematorium, the device sits next to a bowl of the statistical average of all slow-cooked meals, a few inches (or feet?) away from a rare leafy bell pepper, in front of some towering greens in a utensil holder, and looks like a miniature; on the bottom right, in the middle of a butcher block, it looks fairly large, except for some confoundingly scaled [robot voice] STARCH and MEAT on the plate.

«

»

Let’s say I am the owner of a hedge fund, and one fine June day, my employees come to me and say, “Hey, Liz, we have an accounting problem. We are missing several billion dollars.” How would I react?

I have been wondering this since Danielle Sassoon walked Sam Bankman-Fried through his reaction to the FTX software bug fixed by Adam Yedidia. In my case, there would probably be shouting? Like, a lot of shouting. I would also probably have my assistant figure out which law enforcement agency to call immediately. Misplacing $900m is a five-alarm fire even for Citibank; misplacing several billion is kicking over a lantern in Chicago in 1871.

Obviously, this is not how Bankman-Fried reacted to the software bug that overestimated the amount Alameda owed to FTX by about $8bn. Nor is it how he reacted to finding out that even after fixing the bug, Alameda still owed FTX about $8bn. Instead, Bankman-Fried directed alleged co-conspirator Caroline Ellison to repay third-party loans and went on making investments. 

Bankman-Fried said that he found out about the fiat@ftx account — that’s the one that tracked how much Alameda owed FTX — in June 2022 while his senior staff was focused on fixing the software bug. He did not, however, discover what the account was for until October. I know this sounds unbelievable, but this is his actual testimony. 

«

»

OpenAI has “discussed and debated quite extensively” when to release a tool that can determine whether an image was made with DALL-E 3, OpenAI’s generative AI art model, or not. But the startup isn’t close to making a decision anytime soon.

That’s according to Sandhini Agarwal, an OpenAI researcher who focuses on safety and policy, who spoke with TechCrunch in a phone interview this week. She said that, while the classifier tool’s accuracy is “really good” — at least by her estimation — it hasn’t met OpenAI’s threshold for quality.

“There’s this question of putting out a tool that’s somewhat unreliable, given that decisions it could make could significantly affect photos, like whether a work is viewed as painted by an artist or inauthentic and misleading,” Agarwal said.

OpenAI’s targeted accuracy for the tool appears to be extraordinarily high. Mira Murati, OpenAI’s chief technology officer, said this week at The Wall Street Journal’s Tech Live conference that the classifier is “99%” reliable at determining if an unmodified photo was generated using DALL-E 3. Perhaps the goal is 100%; Agarwal wouldn’t say.

A draft OpenAI blog post shared with TechCrunch revealed this interesting tidbit:

“[The classifier] remains over 95% accurate when [an] image has been subject to common types of modifications, such as cropping, resizing, JPEG compression, or when text or cutouts from real images are superimposed onto small portions of the generated image.”

«

»

“Word templates led people to use the same formatting in communications, and eventually, this has become instantiated as a norm,” says Gloria Mark, a professor of informatics at the University of California, Irvine, where she studies human-computer interaction. If you work in finance, there’s a specific way reports are expected to be laid out. Letters follow a set pattern, memos are largely formatted in the same way. “Users know where to find information in these standardised documents; they don’t need to spend time trying to find what they need.”

If you take this idea of professional conformity a step further, Word has also been significant in helping establish English as the global language of business. While it would be an overstatement to say Word alone made English the dominant language, as a US firm, Microsoft’s mother-tongue is American-English. When this is coupled with Word’s ubiquity, it at least reinforces this dominance.

“Word primarily operates in English,” says Noël Wolf, a linguistic expert at the language learning platform Babbel. “As businesses become increasingly global, the widespread use of Word in professional and technical fields has led to the borrowing of English terms and structures, which contribute to the trend of linguistic homogenisation.”

Word’s spell-checker and grammar features have become subtle arbiters of language, too. Although seemingly trivial, these tools “promote a sense of consistency and correctness”, says Wolf, and this uniformity comes at the cost of writing diversity.  “Writers, when prompted by the software’s automated norms, might unintentionally forsake their unique voices and expressions.”

This becomes even more invasive when you look at the role and impact of autocorrect and predictive text. Today, when typing on Word, the software can automatically correct your spelling, and make suggestions for what to write next. These suggestions aren’t (yet) based on your personal writing style and tone – they’re rule-based. The suggestions you see will be the same as millions of others. Again, this may feel innocuous but it’s another example of how Word standardises language by loosely guiding everyone down the same path. 

«

»

Rental car company Hertz once envisioned itself as the ultimate EV broker, doling out battery-powered vehicles to business travelers, ridehail drivers, and tech newbies in an ambitious plan to grease the wheels for the EV revolution. The company inked agreements with Tesla and Polestar to buy nearly 200,000 EVs. Tesla’s valuation topped $1 trillion on the news.

But like many other aspects of the EV switch, that plan is running into some headwinds. This week, Hertz said it was tapping the brakes on its EV rollout, citing the plummeting resale value of its EVs and the high cost of repair.

Tesla has been slashing prices to spark sales as it finds itself wrestling with softening demand and more competition. And repair costs are about double what the company spends on gas car fixes, Hertz CEO Stephen Scherr told Bloomberg.

Part of the problem is linked to Hertz’s plans to rent EVs to ridehail drivers. Of the 100,000 Tesla acquired by Hertz, half were to be allocated to Uber drivers as part of a deal with the ridehail company. And drivers said they loved the Teslas! But Uber drivers also tend to drive their vehicles into the ground. This higher rate of utilization can lead to a lot of damage — certainly more than Hertz was anticipating.

Hertz said it tried to mitigate “higher incidents of damage among EV rideshare drivers” by siphoning some of its fleet into its leisure segment. But that didn’t work out as well as the company hoped.

…Bloomberg says that factoring out its EV costs, Hertz probably would have met Wall Street’s expectations for the quarter. The company’s stock dropped about 15% in the past five days.

«

»

In an internal memo shared with Apple Authorized Service Providers, Apple said charging an iPhone 15, iPhone 15 Plus, iPhone 15 Pro, or iPhone 15 Pro Max with a “small number” of wireless phone chargers built into certain recent BMW and Toyota Supra models may temporarily disable the NFC capabilities of the device. The memo does not provide a specific reason for the issue, which we first reported earlier this month.

In an iPhone, the NFC chip powers features like Apple Pay and digital car keys. Users affected by this issue may receive a “Could Not Set Up Apple Pay” error message in the Wallet app, even if they have already set up Apple Pay.

Apple said the issue will be addressed in a software update coming later this year. Given the memo was put out hours after iOS 17.1 was released today, it appears the update will require a subsequent update, such as iOS 17.1.1 or iOS 17.2. In the meantime, Apple advises affected customers to stop using in-car wireless charging.

There are some complaints about this issue across the MacRumors Forums and X, formerly known as Twitter, but it is unclear how widespread the problem is. Affected customers said the iPhone goes into a data recovery mode with a white screen, and the NFC chip is no longer functional after the device reboots. There is no way for customers to fix the problem, so a trip to an Apple Store or an Apple Authorized Service Provider is required.

«

»

Andrew Ng, a professor at Stanford University who taught machine learning to the likes of OpenAI co-founder Sam Altman, and who himself co-founded Google Brain and was chief scientist at Baidu’s Artificial Intelligence Group, said that the “bad idea that AI could make us go extinct” was merging with the “bad idea that a good way to make AI safer is to impose burdensome licensing requirements” on the AI industry.

“When you put those two bad ideas together, you get the massively, colossally dumb idea [of] policy proposals that try to require licensing of AI,” Professor Ng told The Australian Financial Review in an interview. “It would crush innovation,” he said.

“There are definitely large tech companies that would rather not have to try to compete with open source [AI], so they’re creating fear of AI leading to human extinction. It’s been a weapon for lobbyists to argue for legislation that would be very damaging to the open-source community,” he said.

In May, OpenAI CEO and co-founder Altman co-signed a letter saying that “mitigating the risk of extinction from AI should be a global priority”, and in March, more than 1100 industry leaders including Elon Musk and Apple co-founder Steve Wozniak called for a six-month moratorium on training powerful AI models.

“Sam [Altman] was one of my students at Stanford. He interned with me. I don’t want to talk about him specifically because I can’t read his mind, but …I feel like there are many large companies that would find it convenient to not have to compete with open-sourced large language models,” he said.

«