AI

OpenAI, DeepMind Will Open Up Models To UK Government (politico.eu) 17

Google DeepMind, OpenAI and Anthropic have agreed to open up their AI models to the U.K. government for research and safety purposes, Prime Minister Rishi Sunak announced at London Tech Week on Monday. From a report: The priority access will be granted in order "to help build better evaluations and help us better understand the opportunities and risks of these systems," Sunak said. The announcement came in a speech that championed the promise of AI to transform areas such as education and healthcare and heralded the U.K.'s potential as an "island of innovation."

"AI is surely one of the greatest opportunities before us," said Sunak. By combining AI models with the power of quantum, "the possibilities are extraordinary," he marvelled. "But we must and we will do it safely," he continued. "I know people are concerned."

Google

Google To Include Office Attendance In Performance Reviews (cnbc.com) 135

Google is implementing stricter measures to enforce office attendance, including tracking badge data, confronting employees who don't come in as required, and factoring attendance into performance reviews. CNBC reports: Google's chief people officer, Fiona Cicconi, wrote an email to employees at the end of the day on Wednesday, which included doubling down on office attendance, reasoning that "there's just no substitute for coming together in person." "Of course, not everyone believes in 'magical hallway conversations,' but there's no question that working together in the same room makes a positive difference," Cicconi's email read. "Many of the products we unveiled at I/O and Google Marketing Live last month were conceived, developed and built by teams working side by side."

Her note said the company will start including their three days per week as a part of their performance reviews and teams will start sending reminders to workers "who are consistently absent from the office." Cicconi even asked already-approved remote workers to reconsider. "For those who are remote and who live near a Google office, we hope you'll consider switching to a hybrid work schedule. Our offices are where you'll be most connected to Google's community." A separate internal document showed that already-approved remote workers may be subject to reevaluation if the company determines "material changes in business need, role, team, structure or location."

In the U.S., the company will periodically track whether employees are adhering to the office attendance policy using badge data, and executives are currently reviewing local requirements to implement in other countries, one of the documents states. If workers don't follow the policy after an extended period of time, human resources will reach out about "next steps." Going forward, Cicconi said, new fully remote work will only be granted "by exception only."
In a statement to CNBC, Google spokesperson Ryan Lamont said, "our hybrid approach is designed to incorporate the best of being together in person with the benefits of working from home for part of the week. Now that we're more than a year into this way of working, we're formally integrating this approach into all of our workplace policies."

Lamont added that the badge data viewed by company leaders is aggregate data and not individualized.
United States

FBI Abused Spy Law 280,000 Times In a Year (theregister.com) 151

The FBI misused surveillance powers granted by Section 702 of the Foreign Intelligence Surveillance Act (FISA) over 278,000 times between 2020 and early 2021 to conduct warrantless searches on George Floyd protesters, January 6 Capitol rioters, and donors to a congressional campaign, according to a newly unclassified court opinion. The Register reports: On Friday, the US Foreign Intelligence Surveillance Court made public a heavily redacted April 2022 opinion [PDF] that details hundreds of thousands of violations of Section 702 of the Foreign Intelligence Surveillance Act (FISA) -- the legislative instrument that allows warrantless snooping. The Feds were found to have abused the spy law in a "persistent and widespread" manner, according to the court, repeatedly failing to adequately justify the need to go through US citizens' communications using a law aimed at foreigners.

The court opinion details FBI queries run on thousands of individuals between 2020 and early 2021. This includes 133 people arrested during the George Floyd protests and more than 19,000 donors to a congressional campaign. In the latter, "the analyst who ran the query advised that the campaign was a target of foreign influence, but NSD determined that only eight identifiers used in the query had sufficient ties to foreign influence activities to comply with the querying standard," the opinion says, referring to the Justice Department's National Security Division (NSD). In other words, there wasn't a strong enough foreign link to fully justify the communications search.

For the Black Lives Matter protests, the division determined that the FBI queries "were not reasonably likely to retrieve foreign intelligence information or evidence of a crime." Again, an overreach of foreign surveillance powers. Additional "significant violations of the querying standard" occurred in searched related to the January 6, 2021 breach of the US Capitol, domestic drug and gang investigations, and domestic terrorism probes, according to the court. It's said that more than 23,000 queries were run on people suspected of storming the Capitol.

China

UC Berkeley Neglected To Disclose $220 Million Deal With China To the US Government (thedailybeast.com) 63

schwit1 shares a report from The Daily Beast: U.C. Berkeley has failed to disclose to the U.S. government massive Chinese state funding for a highly sensitive $240 million joint tech venture in China that has been running for the last eight years. The Californian university has not registered with the U.S. government that it received huge financial support from the city of Shenzhen for a tech project inside China, which also included partnerships with Chinese companies that have since been sanctioned by the U.S. or accused of complicity in human rights abuses.

The university has failed to declare a $220 million investment from the municipal government of Shenzhen to build a research campus in China. A Berkeley spokesperson told The Daily Beast that the university had yet to declare the investment -- announced in 2018 -- because the campus is still under construction. However, a former Department of Education official who used to help manage the department's foreign gifts and contracts disclosure program said that investment agreements must be disclosed within six months of signing, not when they are fully executed. Berkeley admitted that it had also failed to disclose to the U.S. government a $19 million contract in 2016 with Tsinghua University, which is controlled by the Chinese government's Ministry of Education.

The project's Chinese backers promised lavish funding, state-of-the-art equipment, and smart Ph.D. students for Berkeley academics researching national security-sensitive technologies, according to contract documents exclusively obtained by The Daily Beast. After the project got underway, Berkeley researchers granted Chinese officials private tours of their cutting-edge U.S. semiconductor facilities and gave "priority commercialization rights" for intellectual properties (IP) they produced to Chinese government-backed funds. A Berkeley spokesman said that Berkeley only pursued fundamental research through TBSI, meaning that all research projects were eventually publicly published and accessible to all; it did not conduct any proprietary research that exclusively benefited a Chinese entity. Still, Berkeley's ties to the Chinese government and sanctioned Chinese companies are sure to raise eyebrows in Washington, where U.S. policymakers are increasingly concerned about the outflow of U.S. technology to China, especially those with military applications.

United States

DOJ Charges Former Apple Engineer With Theft of Autonomous Car Tech for China (cnbc.com) 30

A former Apple software engineer was charged with stealing Apple's autonomous technology for a Chinese self-driving car company, the Department of Justice announced Tuesday. From a report: Weibao Wang worked as a software engineer at Apple from 2016 to 2018, a DOJ indictment said. Wang worked on Apple's Annotation Team and was granted "broad access" to databases which the Justice Department said could only be accessed by 2,700 of Apple's 135,000 employees. Wang is the third former Apple employee to be accused of stealing autonomous trade secrets for China.

Wang has been charged with six separate counts involving the theft or attempted theft of Apple's "entire autonomy source code," tracking systems, behavior planning for autonomous systems, and descriptions of the hardware that was behind the systems. A year into his employment, four months before he quit his job at Apple, Wang accepted a job at the U.S.-based subsidiary of an unnamed Chinese company that was developing autonomous driving technology and he began to siphon "large amounts" of sensitive commercial technology and source code, the indictment alleges.

The Courts

Google Gets Court Order To Take Down CryptBot That Infected Over 670,000 Computers (thehackernews.com) 14

An anonymous reader quotes a report from The Hacker News: Google on Wednesday said it obtained a temporary court order in the U.S. to disrupt the distribution of a Windows-based information-stealing malware called CryptBot and "decelerate" its growth. The tech giant's Mike Trinh and Pierre-Marc Bureau said the efforts are part of steps it takes to "not only hold criminal operators of malware accountable, but also those who profit from its distribution." CryptBot is estimated to have infected over 670,000 computers in 2022 with the goal of stealing sensitive data such as authentication credentials, social media account logins, and cryptocurrency wallets from users of Google Chrome. The harvested data is then exfiltrated to the threat actors, who then sell the data to other attackers for use in data breach campaigns. CryptBot was first discovered in the wild in December 2019.

The malware has been traditionally delivered via maliciously modified versions of legitimate and popular software packages such as Google Earth Pro and Google Chrome that are hosted on fake websites. [...] The major distributors of CryptBot, per Google, are suspected to be operating a "worldwide criminal enterprise" based out of Pakistan. Google said it intends to use the court order, granted by a federal judge in the Southern District of New York, to "take down current and future domains that are tied to the distribution of CryptBot," thereby kneecapping the spread of new infections.

Bitcoin

US Crypto Exchange Coinbase Secures Bermuda License (reuters.com) 16

Coinbase has been granted a license by the Bermuda Monetary Authority, allowing the US crypto exchange to operate as a digital asset business there. The exchange is also in the process of obtaining a license in Abu Dhabi. Reuters reports: Coinbase CEO Brian Armstrong said on Tuesday that crypto firms will develop in "offshore" havens unless the U.S. and UK create "clarity about regulation" for crypto. Coinbase is planning to launch a crypto derivatives exchange in Bermuda as soon as next week, Fortune reported on Wednesday, citing a person close to the company.

U.S. SEC Chair Gary Gensler told lawmakers on Tuesday that he had "never seen a field that's so non-complying with laws." Crypto firms say they need clarity about regulations, but Gensler has said that crypto markets "suffer from a lack of regulatory compliance, not a lack of regulatory clarity."

Links

Man Battling Google Wins $500K For Search Result Links Calling Him a Pedophile (arstechnica.com) 32

An anonymous reader quotes a report from Ars Technica: A Montreal man spent years trying to hold Google accountable for search results linking to a defamatory post falsely accusing him of pedophilia that he said ruined his career. Now Google must pay $500,000 after a Quebec Supreme Court judge ruled that Google relied on an "erroneous" interpretation of Canadian law in denying the man's requests to remove the links. "Google variously ignored the Plaintiff, told him it could do nothing, told him it could remove the hyperlink on the Canadian version of its search engine but not the US one, but then allowed it to re-appear on the Canadian version after a 2011 judgment of the Supreme Court of Canada in an unrelated matter involving the publication of hyperlinks," judge Azimuddin Hussain wrote in his decision (PDF) issued on March 28.

The plaintiff was granted anonymity throughout the proceedings. Google has been ordered not to disclose any identifiable information about him in connection to the case for 45 days. The tech company must also remove all links to the defamatory post in search results viewable in Quebec. [...] Instead of compensatory and punitive damages originally sought -- amounting to $6 million -- the man was awarded $500,000 for moral injuries caused after successfully arguing that he lost business deals and suffered strains on his personal relationships due to being wrongly stigmatized as a pedophile. Hussain described the plaintiff's experience battling Google to preserve his reputation as a "waking nightmare." Due to Google's refusals to remove the defamatory posts, the man "found himself helpless in a surreal and excruciating contemporary online ecosystem as he lived through a dark odyssey to have the Defamatory Post removed from public circulation," Hussain wrote. The plaintiff, now in his early 70s, has the option to appeal the judge's order that Google may not release any of his identifiable information for 45 days.

Build

The Orange Pi 5: a Fast Alternative To The Raspberry Pi 4 (phoronix.com) 81

"With an 8-core Rockchip RK3588S SoC, the Orange Pi 5 is leaps and bounds faster than the aging Raspberry Pi 4," writes Phoronix: With up to 32GB of RAM, the Orange Pi 5 is also capable of serving for a more diverse user-base and even has enough potential for assembling a budget Arm Linux developer desktop. I've been testing out the Orange Pi 5 the past few weeks and it's quite fast and nice for its low price point.

The Orange Pi 5 single board computer was announced last year and went up for pre-ordering at the end of 2022.... When it comes to the software support, among the officially available options for the Orange Pi 5 are Orange Pi OS, Ubuntu, Debian, Android, and Armbian. Other ARM Linux distributions will surely see varying levels of support while even the readily available ISO selection offered by Orange Pi is off to a great start....

Granted, the Orange Pi developer community isn't as large as that of the Raspberry Pi community or the current range of accessories and documentation, but for those more concerned about features and performance, the Orange Pi 5 is extremely interesting.

The article includes Orange Pi 5 specs:
  • A 26-pin header
  • HDMI 2.1, Gigabit LAN, M.2 PCIe 2.0, and USB3 connectivity
  • A Mali G510 MP4 graphics processor, "which has open-source driver hope via the Panfrost driver stack."
  • Four different versions with 4GB, 8GB, 16GB, or 32GB of RAM using LPDDR4 or LPDDR4X. "The Orange Pi 4GB retails for ~$88, the Orange Pi 5 8GB version retails for $108, and the Orange Pi 5 16GB version retails for $138, while as of writing the 32GB version wasn't in stock."

In 169 performance benchmarks (compared to Raspberry Pi 4 boards), "this single board computer came out to delivering 2.85x the performance of the Raspberry Pi 400 overall." And through all this the average SoC temperature was 71 degrees with a peak of 85 degrees — without any extra heatsink or cooling.


United States

The Spy Law That Big Tech Wants To Limit (bloomberg.com) 26

Top tech companies are mounting a push to limit how US intelligence agencies collect and view texts, emails and other information about their users, especially American citizens. From a report: The companies, including Alphabet's Google, Meta Platforms and Apple, want Congress to limit Section 702 of the Foreign Intelligence Surveillance Act, as they work to renew the law before it expires at the end of the year, according to three people familiar with the discussions. There is a growing bipartisan consensus in Congress to not only renew the law but to make changes in response to a series of reports and internal audits documenting abuses. That's left the tech industry optimistic that broader reforms will get through Congress this time, according to two lobbyists who asked not to be identified relaying internal discussions.

The law, passed by Congress in 2008 in response to revelations of warrantless spying on US citizens by the Bush administration, granted sweeping powers that have been criticized over the years for different reasons. Civil liberties groups think more privacy protections are needed. Former President Donald Trump and his allies claim that spying powers enable intelligence agencies to conspire against conservatives. "Reforms are needed to ensure dragnet surveillance programs operate within constitutional limits and safeguard American users' rights, through appropriate transparency, oversight and accountability," said Matt Schruers, president of the tech trade group Computer & Communications Industry Association, which counts Apple, Google, Meta and Amazon among its members. Intelligence agencies say Section 702 is an essential tool that has generated critical information on the espionage and hacking activities of countries such as China and contributed to the successful drone strike that killed al-Qaeda leader Ayman al-Zawahiri last year.

Medicine

2 Drug Companies Can Legally Start Selling Cocaine, Heroin, and MDMA (vice.com) 179

An anonymous reader quotes a report from Motherboard: At least two companies in British Columbia, Canada, say they've received exemptions from the federal government allowing them to produce and distribute cocaine, heroin, MDMA, or magic mushrooms. But it's not clear under what circumstances the companies will be able to sell the drugs, and B.C. Premier David Eby said he was "astonished" to hear the announcement. On Thursday, Sunshine Earth Labs, a psychedelics manufacturer announced that Health Canada, a federal government agency, is allowing the company to legally produce and distribute the coca leaf and cocaine; MDMA; opium; morphine, heroin and psilocybin, the active ingredient in shrooms. The company said it plans to "bring a safer supply of drugs to the global market."

Meanwhile, cannabis extractions company Adastra announced it's now legally allowed to both produce and distribute psilocybin and cocaine. In a statement to VICE News, Health Canada said Adastra is licensed to produce the drugs for scientific and medical purposes but cannot sell products to the general public. "They are only permitted for sale to other licence holders who have cocaine listed on their licence, pharmacists, practitioners, hospitals, or the holder of a section 56(1) exemption for research purposes," the agency said.

Both companies claim they received amendments under Health Canada's Dealer's Licenses, which grant manufacturers, doctors, and researchers exemptions to the Controlled Drugs and Substances Act, allowing them to legally possess and make banned drugs. In a news conference, Eby said the licenses were granted without consultation from the province. "It is not part of our provincial plan," he said, noting that he would be following up with Health Canada about the announcements. Adastra said it's license allows it to "interact with up to 250 grams of cocaine and to import coca leaves to manufacture and synthesize the substance."

Patents

Dell and Partners Smash Patent Troll WSOU in Court (beehiiv.com) 37

In the land of patent litigation, all patent trolls want to file in the US Western District of Texas Court. This court is infamous for being sympathetic to patent plaintiffs. That's why patent litigator WSOU Investments, aka Brazos Licensing and Development, went after Dell, EMC, and VMware in this Court. Usually, this would have been the smart move. Not this time. District Judge Alan Albright granted the defendants a directed verdict, and that was the end of the matter. From a report: What happened was this: WSOU, although successful before with their carpet bombing patent lawsuit strategy, failed this time. According to the lead defense counsel and Gibson Dunn partner, Brian A. Rosenthal, "This case got to trial because the plaintiff refused to come to their senses before trial. We obtained a number of serious exclusions of evidence prior to trial, and told them very early on the case had no merit." The judge agreed.

That came as a surprise to those of us who watch patent lawsuits, so you don't have to. As Heather Meeker, the well-known open-source and intellectual property (IP) lawyer, said, "This is surprisingly defendant-friendly from Judge Albright, who has received a lot of criticism for making Waco such a patent plaintiff-friendly docket." Until now, WSOU had been very successful. As a Patent Assertion Entity (PAEs), its only goal is to profit by acquiring patents and then suing companies that might be using the patents' intellectual property (IP) assets. It does this by using its portfolio of technology patents to file numerous individual suits involving different patents against companies. WSOU's main tactic, as Unified Patents put it, "forces operating companies to either settle or fight, on average, eight lawsuits at once."

Most companies faced with the financial burden of struggling with so many lawsuits settle rather than fight. Not this time. For the first time, companies decided to take the issues to court. In this particular set of cases, WSOU claimed in a June 2020 lawsuit that the defendants had infringed on three cloud infrastructure networking patents, and sought $435 million in damages. Rosenthal argued that the patents in question were old and irrelevant to the defendants' interests. The defense team had informed WSOU in October 2020 that there was no proof of direct infringement, but the plaintiff persisted with the case, leading to exclusions of evidence prior to trial. So it was that on the first day of the trial, two of the patents were tossed out on evidentiary rulings, and the plaintiff rested its case on the third day. The defense then requested a directed verdict, which was granted by Albright, resulting in a win for the defendants. In short, even this patent-friendly court could find no evidence at all for WSOU's assertions.

Facebook

Meta Has a New Machine Learning Language Model (theverge.com) 16

The buzz in tech these last few weeks has been focused squarely on the language models developed and deployed by the likes of Microsoft, Google, and OpenAI. But Meta, Facebook's parent company, continues to do significant work in this field and is releasing a new AI language generator named LLaMA today. From a report: LLaMA isn't like ChatGPT or Bing; it's not a system that anyone can talk to. Rather, it's a research tool that Meta says it's sharing in the hope of "democratizing access in this important, fast-changing field." In other words: to help experts tease out the problems of AI language models, from bias and toxicity to their tendency to simply make up information.

To this end, Meta is releasing LLaMA (which is not actually a single system but a quartet of different-sized models) under "a noncommercial license focused on research use cases," with access granted to groups like universities, NGOs, and industry labs. "We believe that the entire AI community -- academic researchers, civil society, policymakers, and industry -- must work together to develop clear guidelines around responsible AI in general and responsible large language models in particular," the company wrote in its post. "We look forward to seeing what the community can learn -- and eventually build -- using LLaMA."

AI

AI-Created Images Lose US Copyrights In Test For New Technology (reuters.com) 100

Images in a graphic novel that were created using the artificial-intelligence system Midjourney should not have been granted copyright protection, the U.S. Copyright Office said in a letter seen by Reuters. From the report: "Zarya of the Dawn" author Kristina Kashtanova is entitled to a copyright for the parts of the book she wrote and arranged, but not for images she made using Midjourney, the office said in its letter, dated Tuesday. The decision is one of the first by a U.S. court or agency on the scope of copyright protection for works created with AI, and comes amid the meteoric rise of generative AI software like Midjourney, Dall-E and ChatGPT.

The Copyright Office said in its letter that it would reissue its registration for "Zarya of the Dawn" to omit images that "are not the product of human authorship." [...] Midjourney is an AI-based system that generates images based on text prompts entered by users. Kashtanova wrote the text of "Zarya of the Dawn," and Midjourney created the book's images based on her prompts.

The Courts

Supreme Court Rejects Ohio Man's Bid To Sue Police Over Arrest of Facebook Parody (nbcnews.com) 83

An anonymous reader quotes a report from NBC News: The Supreme Court on Tuesday turned away an Ohio man's claim that his constitutional rights were violated when he was arrested and prosecuted for making satirical posts about his local police department on Facebook. The justices' rejection of Anthony Novak's appeal means his civil rights lawsuit against the Parma Police Department cannot move forward. With its decision, the court again declined to consider revisiting "qualified immunity," the contentious legal defense that lets police officers and other government officials off the hook in civil rights cases if constitutional violations have not been "clearly established" when they occur. At issue was whether a lower court correctly granted the police officers qualified immunity under the rationale that previous court precedent had not clearly established that Novak's actions constituted protected speech under the Constitution's First Amendment.

In March 2016, Novak set up a Facebook page that purported to be that of the Parma Police Department. He published six satirical posts in 12 hours, one of which claimed there was a job opening to which minorities were encouraged not to apply and another that warned people not to give food, money or shelter to homeless people. The police department, claiming the posts had disrupted its operations, launched an investigation and ultimately searched Novak's apartment, arrested him and jailed him for four days. Novak was charged under a state law that criminalizes disruption of police operations but acquitted at trial.

The police officers, Kevin Riley and Thomas Connor, say they had probable cause to arrest Novak because they genuinely believed his conduct was disrupting their operations. Novak sued the officers and the police department, saying they had violated his free speech rights, as well as his right to be free of unlawful searches and seizures under the Constitution's Fourth Amendment. After lengthy litigation, a federal judge dismissed Novak's claims. The Cincinnati-based 6th U.S. Circuit Court of Appeals agreed in a ruling in April that "the officers reasonably believed they were acting within the law" even if his Facebook page was obviously a parody. That's because there was no court precedent saying it's a violation of the Constitution to be arrested in retaliation for satirical remarks when the officers have probable cause, the court said.
Novak's appeal was backed by satirical news sites The Babylon Bee and The Onion, which filed a lighthearted brief saying its writers "have a self-serving interest in preventing political authorities from imprisoning humorists."
Businesses

Zantac's Maker Kept Quiet About Cancer Risks for 40 Years (bloomberg.com) 98

Glaxo says the heartburn drug doesn't cause tumors. But the company was warned by its own scientists and independent researchers about the potential danger. From a report: The small British company was sometimes called Glaxo University, because it conducted important pharmaceutical research that rarely resulted in profitable drugs. Then the scientists at Glaxo Laboratories created a molecule they called ranitidine, and in 1978 the company was granted a US patent. The molecule was new, but not novel. The scientists had, as scientists sometimes do, looked for a way to mimic the success of an established drug -- in this case, one that healed ulcers and could be used to treat heartburn. They developed ranitidine quickly, and the US Food and Drug Administration reviewed it quickly. Glaxo gave it the brand name Zantac. Glaxo marketed it as better and safer than the drug that inspired it, Tagamet, and before long, Zantac overtook Tagamet to become the world's bestselling prescription medication. For years, Glaxo counted on Zantac for nearly half of its sales and almost as much of its profit. The company won an award from Queen Elizabeth; the chief executive officer was knighted. Zantac created reputations and fortunes. It financed the modern version of Glaxo, which, after mergers and takeovers and spinoffs, ended up as GSK, a company now worth some $73 billion. Among its most popular drugs are the antidepressants Paxil and Wellbutrin and the shingles vaccine Shingrix.

But not Zantac. In 2019 the drug was found to be tainted with high levels of a probable carcinogen. Not by chance or mistake in a few batches. The poison is created by ranitidine itself. Zantac's makers and health regulators around the world recalled the drug, and in the spring of 2020 the FDA forced it off the market altogether. No company could manufacture it; nobody should ingest it. The carcinogen, called NDMA, was once added to rocket fuel and is now used only to induce cancer in lab rats. The FDA says consuming minuscule amounts isn't harmful. But tests were revealing excessive amounts of NDMA in ranitidine -- and a capacity to create even more over time. No version seemed safe. From ranitidine's beginning to its end, Glaxo had been warned by its own scientists and independent researchers about the potential danger. An account of those four decades emerges in hundreds of documents, thousands of pages, many of which have never been made public. Bloomberg Businessweek reviewed court filings, many still under seal, as well as studies, FDA transcripts and new drug applications obtained via Freedom of Information Act requests. They show that the FDA considered the cancer risks when approving ranitidine. But Glaxo didn't share a critical study. Over the years, the company also backed flawed research designed to minimize concerns and chose not to routinely transport and store the medication in ways that could have eased the problem. Glaxo sold a drug that might harm people, tried to discount evidence of that and never gave anyone the slightest warning. More than 70,000 people who took Zantac or generic versions of it are suing the company in US state courts for selling a potentially contaminated and dangerous drug.

Power

EverWind Gets Approval For North America's First Green Hydrogen Facility (reuters.com) 62

EverWind Fuels has become the first green hydrogen producer in North America to secure the necessary permits for a commercial-scale facility on Tuesday. Reuters reports: Provincial authorities in Canada granted environmental approval for EverWind to begin converting a former oil storage facility and marine terminal at Point Tupper in Nova Scotia into a green hydrogen and ammonia production hub. [...] EverWind expects the project's first phase, producing and exporting 200,000 tonnes per annum, to be online in 2025, before ramping up to 1 million tonnes per annum the following year. The company has agreements with German energy firms E.ON and Uniper to acquire the production. "To get the permit is a big deal," said Vichie, who co-founded Stonepeak Infrastructure Partners and also worked at Blackstone. The green hydrogen produced by EverWind's facility will be combined with nitrogen and converted into ammonia before being shipped, in liquid form, in tankers to Germany, where it can be retained as ammonia or turned back into green hydrogen.

Production during the facility's first phase will be powered using wind and solar assets to be built nearby, Vichie said. The company in December leased 137,000 acres (55,440 hectares) which will eventually site turbines generating 2 gigawatts of wind energy, that will power production in its second, larger phase. "This provides an amazing green growth path for Atlantic Canada, where they have some of the world's best wind resources," Vichie said. The overall cost of the project is expected to be around $6 billion. Three banks are helping arrange debt funding, while Vichie's family office is providing equity capital, he said.
In other hydrogen-related news, the Federal Aviation Administration (FAA) granted Universal Hydrogen approval to fly its 40+ passenger hydrogen electric plane.
Android

Bloatware Pushes the Galaxy S23 Android OS To an Incredible 60GB (arstechnica.com) 92

An anonymous reader quotes a report from Ars Technica: As a smartphone operating system, Android strives to be a lightweight OS so it can run on a variety of hardware. The first version of the OS had to squeeze into the T-Mobile G1, with only a measly 256MB of internal storage for Android and all your apps, and ever since then, the idea has been to use as few resources as possible. Unless you have the latest Samsung phone, where Android somehow takes up an incredible 60GB of storage. Yes, the Galaxy S23 is slowly trickling out to the masses, and, as Esper's senior technical editor Mishaal Rahman highlights in a storage space survey, Samsung's new phone is way out of line with most of the ecosystem. Several users report the phone uses around 60GB for the system partition right out of the box. If you have a 128GB phone, that's nearly half your storage for the Android OS and packed-in apps. That's four times the size of the normal Pixel 7 Pro system partition, which is 15GB. It's the size of two Windows 11 installs, side by side. What could Samsung possibly be putting in there?!

We can take a few guesses as to why things are so big. First, Samsung is notorious for having a shoddy software division that pumps out low-quality code. The company tends to change everything in Android just for change's sake, and it's hard to imagine those changes are very good. Second, Samsung may want to give the appearance of having its own non-Google ecosystem, and to do that, it clones every Google app that comes with its devices. Samsung is contractually obligated to include the Google apps, so you get both the Google and Samsung versions. That means two app stores, two browsers, two voice assistants, two text messaging apps, two keyboard apps, and on and on. These all get added to the system partition and often aren't removable.

Unlike the clean OSes you'd get from Google or Apple, Samsung sells space in its devices to the highest bidder via pre-installed crapware. A company like Facebook will buy a spot on Samsung's system partition, where it can get more intrusive system permissions that aren't granted to app store apps, letting it more effectively spy on users. You'll also usually find Netflix, Microsoft Office, Spotify, Linkedin, and who knows what else. Another round of crapware will also be included if you buy a phone from a carrier, i.e., all the Verizon apps and whatever space they want to sell to third parties. The average amount users are reporting is 60GB, but crapware deals change across carriers and countries, so it will be different for everyone.

China

US Stops Granting Export Licenses For China's Huawei (reuters.com) 36

An anonymous reader quotes a report from Reuters: The Biden administration has stopped approving licenses for U.S. companies to export most items to China's Huawei, according to three people familiar with the matter. Huawei has faced U.S. export restrictions around items for 5G and other technologies for several years, but officials in the U.S. Department of Commerce have granted licenses for some American firms to sell certain goods and technologies to the company. Qualcomm in 2020 received permission to sell 4G smartphone chips to Huawei.

One person familiar with the matter said U.S. officials are creating a new formal policy of denial for shipping items to Huawei that would include items below the 5G level, including 4G items, Wifi 6 and 7, artificial intelligence, and high-performance computing and cloud items. Another person said the move was expected to reflect the Biden administration's tightening of policy on Huawei over the past year. Licenses for 4G chips that could not be used for 5g, which might have been approved earlier, were being denied, the person said. Toward the end of the Trump administration and early in the Biden administration, officials had still granted licenses for items specific to 4G applications.
Chinese foreign ministry spokesperson Mao Ning said that China opposes the United States abusing an overly broad notion of national security to suppress Chinese firms unreasonably. The move "goes against the principles of the market economy and rules of international trade and finance, hurts the confidence the international community has in the U.S business environment and is blatant technological hegemony," Mao said during a press conference in Beijing on Tuesday.
Biotech

A Drug Company Made $114 Billion Gaming America's Patent System (msn.com) 92

The New York Times looks at the AbbVie's anti-inflammatory drug Humira and their "savvy but legal exploitation of the U.S. patent system." Though AbbVie's patent was supposed to expire in 2016, since then it's maintained a monopoly that generated $114 billion in revenue by using "a formidable wall of intellectual property protection and suing would-be competitors before settling with them to delay their product launches until this year." AbbVie did not invent these patent-prolonging strategies; companies like Bristol Myers Squibb and AstraZeneca have deployed similar tactics to maximize profits on drugs for the treatment of cancer, anxiety and heartburn. But AbbVie's success with Humira stands out even in an industry adept at manipulating the U.S. intellectual-property regime.... AbbVie and its affiliates have applied for 311 patents, of which 165 have been granted, related to Humira, according to the Initiative for Medicines, Access and Knowledge, which tracks drug patents. A vast majority were filed after Humira was on the market.

Some of Humira's patents covered innovations that benefited patients, like a formulation of the drug that reduced the pain from injections. But many of them simply elaborated on previous patents. For example, an early Humira patent, which expired in 2016, claimed that the drug could treat a condition known as ankylosing spondylitis, a type of arthritis that causes inflammation in the joints, among other diseases. In 2014, AbbVie applied for another patent for a method of treating ankylosing spondylitis with a specific dosing of 40 milligrams of Humira. The application was approved, adding 11 years of patent protection beyond 2016.

AbbVie has been aggressive about suing rivals that have tried to introduce biosimilar versions of Humira. In 2016, with Amgen's copycat product on the verge of winning regulatory approval, AbbVie sued Amgen, alleging that it was violating 10 of its patents. Amgen argued that most of AbbVie's patents were invalid, but the two sides reached a settlement in which Amgen agreed not to begin selling its drug until 2023.

Over the next five years, AbbVie reached similar settlements with nine other manufacturers seeking to launch their own versions of Humira. All of them agreed to delay their market entry until 2023.

A drug pricing expert at Washington University in St. Louis tells the New York Times that AbbVie and its strategy with Humira "showed other companies what it was possible to do."

But the article concludes that last year such tactics "became a rallying cry" for U.S. lawmakers "as they successfully pushed for Medicare to have greater control over the price of widely used drugs that, like Humira, have been on the market for many years but still lack competition."

Slashdot Top Deals