Президент Франции Эммануэль Макрон не обладает достаточной квалификацией, чтобы руководить государством, заявил турецкий лидер Реджеп Тайип Эрдоган.
По словам президента Турции, у него состоялось много встреч с Макроном на международном уровне, но его французский коллега был "не честен".
Макрон как президент потерял репутацию в парламенте, а его страна теряет свой престиж в международном сообществе, заключил Эрдоган.
Oracle’s 2023 per-employee pricing for standard Java is raising concerns about its potential impact on Java licensing costs for customers. The pricing is based on total employee counts, not the number of employees using Java.
Published January 23, Oracle’s price list covers the new Java SE Universal Subscription program. The pricing starts at $15 per employee per month for as many as 999 employees, and drops as low as $5.25 per employee per month for 40,000 to 49,999 users. Oracle cited an example in which a company with a total employee count of 28,000, including full-time and part-time employees and agents, consultants, and contractors, would be charged $2.268 million per year.
Google researchers have made an AI that can generate minutes-long musical pieces from text prompts, and can even transform a whistled or hummed melody into other instruments, similar to how systems like DALL-E generate images from written prompts (via TechCrunch). The model is called MusicLM, and while you can’t play around with it for yourself, the company has uploaded a bunch of samples that it produced using the model.
The examples are impressive. There are 30-second snippets of what sound like actual songs created from paragraph-long descriptions that prescribe a genre, vibe, and even specific instruments, as well as five-minute-long pieces generated from one or two words like “melodic techno.” Perhaps my favorite is a demo of “story mode,” where the model is basically given a script to morph between prompts. For example, this prompt:
electronic song played in a videogame (0:00-0:15)
meditation song played next to a river (0:15-0:30)
fire (0:30-0:45)
fireworks (0:45-0:60)
Resulted in this audio:
It may not be for everyone, but I could totally see this being composed by a human (I also listened to it on loop dozens of times while writing this article). Also featured on the demo site are examples of what the model produces when asked to generate 10-second clips of instruments like the cello or maracas (the later example is one where the system does a relatively poor job), eight-second clips of a certain genre, music that would fit a prison escape, and even what a beginner piano player would sound like versus an advanced one. It also includes interpretations of phrases like “futuristic club” and “accordion death metal.”
MusicLM can even simulate human vocals, and while it seems to get the tone and overall sound of voices right, there’s a quality to them that’s definitely off. The best way I can describe it is that they sound grainy or staticky. That quality isn’t as clear in the example above, but I think this one illustrates it pretty well:
That, by the way, is the result of asking it to make music that would play at a gym. You may also have noticed that the lyrics are nonsense, but in a way that you may not necessarily catch if you’re not paying attention — kind of like if you were listening to someone singing in Simlish or that one song that’s meant to sound like English but isn’t.
I won’t pretend to know how Google achieved these results, but it’s released a research paper explaining it in detail if you’re the type of person who would understand this figure:
/cdn.vox-cdn.com/uploads/chorus_asset/file/24391491/Screenshot_2023_01_27_at_15.17.40.png)
Chart: Google
AI-generated music has a long history dating back decades; there are systems that have been credited with composing pop songs, copying Bach better than a human could in the 90s, and accompanying live performances. One recent version uses AI image generation engine StableDiffusion to turn text prompts into spectrograms that are then turned into music. The paper says that MusicLM can outperform other systems in terms of its “quality and adherence to the caption,” as well as the fact that it can take in audio and copy the melody.
That last part is perhaps one of the coolest demos the researchers put out. The site lets you play the input audio, where someone hums or whistles a tune, then lets you hear how the model reproduces it as an electronic synth lead, string quartet, guitar solo, etc. From the examples I listened to, it manages the task very well.
Like with other forays into this type of AI, Google is being significantly more cautious with MusicLM than some of its peers may be with similar tech. “We have no plans to release models at this point,” concludes the paper, citing risks of “potential misappropriation of creative content” (read: plagiarism) and potential cultural appropriation or misrepresentation.
It’s always possible the tech could show up in one of Google’s fun musical experiments at some point, but for now, the only people who will be able to make use of the research are other people building musical AI systems. Google says it’s publicly releasing a dataset with around 5,500 music-text pairs, which could help when training and evaluating other musical AIs.
Взято отсюда
There's plenty of concern that OpenAI's ChatGPT could help students cheat on tests, but just how well would the chatbot fare if you asked it to write a graduate-level exam? It would pass — if only just. In a newly published study, University of Minnesota law professors had ChatGPT produce answers for graduate exams at four courses in their school. The AI passed all four, but with an average grade of C+. In another recent paper, Wharton School of Business professor Christian Terwiesch found that ChatGPT passed a business management exam with a B to B- grade. You wouldn't want to use the technology to impress academics, then.
The research teams found the AI to be inconsistent, to put it mildly. The University of Minnesota group noted that ChatGPT was good at addressing "basic legal rules" and summarizing doctrines, but floundered when trying to pinpoint issues relevant to a case. Terwiesch said the generator was "amazing" with simple operations management and process analysis questions, but couldn't handle advanced process questions. It even made mistakes with 6th grade-level math.
There's room for improvement. The Minnesota professors said they didn't adapt text generation prompts to specific courses or questions, and believed students could get better results with customization. At Wharton, Terwiesch said the bot was adept at changing answers in response to human coaching. ChatGPT might not ace an exam or essay by itself, but a cheater could have the system generate rough answers and refine them.
Both camps warned that schools should limit the use of technology to prevent ChatGPT-based cheating. They also recommended altering the questions to either discourage AI use (such as focusing on analysis rather than reciting rules) or increase the challenge for those people leaning on AI. Students still need to learn "fundamental skills" rather than leaning on a bot for help, the University of Minnesota said.
The study groups still believed that ChatGPT could have a place in the classroom. Professors could teach pupils how to rely on AI in the workplace, or even use it to write and grade exams. The tech could ultimately save time that could be spent on the students, Terwiesch explains, such as more student meetings and new course material.
Взято отсюдаThe U.S. Embassy and Consulates in Mexico have issued a new warning to visitors of Cancun and the Mexican state of Quintana Roo over a potential conflict between taxi drivers and ride-hailing service companies.
Medallion taxi drivers have been attacking ride-hailing service drivers as they battle over lucrative tourist rides. Unfortunately, passengers have been caught in the middle.
“Official complaints against Uber and other drivers do occur … and past disputes between these services and local taxi unions have occasionally turned violent, resulting in injuries to U.S. citizens in some instances,” the warning reads.
On Monday, taxi drivers blocked Kukulcán Boulevard — one of the main roads to the hotel district in Cancun — forcing some passengers to walk or even get police escorts.
Taxi drivers are reportedly furious over a court ruling earlier this month that allows Uber to operate in Quintana Roo without a public transport license. In 2021, the Mexican Supreme Court ruled that Uber is not a public transportation company, so it doesn’t have to comply with the same regulations as taxi drivers. This infuriated taxi drivers, who face higher costs because they are considered public transportation.
Several social media videos show confrontations between taxi drivers and ride-hailing service drivers and their passengers.
It’s not the first time this kind of transportation competition has led to protests. In 2016, there were nationwide strikes in France by taxi unions over the entry of app-based ride services into the market. Those protests turned violent and effectively shut down Paris.
More recently, taxi protests against ride-hailing services like Uber shut down parts of Barcelona.
There have also been large-scale protests against app-based, ride-hailing companies in Belgium, Italy, and several U.S. cities, including Denver.
If you are headed to tourist areas of Mexico, there are other options to get around. The Cancun Sun has some good suggestions.
For example, if you are looking to get around inside the tourist zone in Cancun, there are cheap buses and shuttles. If you’re heading to the Hotel Zone, you can book a private car or van via Expedia or Viator for as little as $30. While it can sometimes be a little pricier, you can also communicate directly with your hotel for the best transportation suggestions at this time.
The State Department had already issued a travel advisory for Mexico based on crime. “In light of widely publicized security incidents in popular tourist destinations, please remember that all destinations have some level of risk,” the advisory reads.
It also suggests visitors review personal security plans, be aware of their surroundings, pay attention to local media and immediately call Mexican 911 in case of any issues.
The Quintana Roo Tourism Board also has a “Guest Assist App” for free download. The app supplies safety information, legal advice, a complaint line, COVID-19 information and access to a 24/7 bilingual call center for visitor assistance.
Гимн Украины представляет собой источник "национальной катастрофы", заявил экс-советник офиса украинского президента Алексей Арестович на своем YouTube-канале.
"Там (в гимне - Ред.) есть очень любопытные строчки. Звучат они так: "души и тела свои положим за свободу". Скажите, в нашем гимне есть хоть какой-нибудь конструктив, какая-нибудь социальная инженерия?" - задался вопросом Арестович.
Он отметил, что из гимна непонятно, что будет после принесения жертв. Кроме того, по мнению Арестовича, формулировка "положим душу и тело" означает войну до момента, пока все не погибнут.Арестович полагает, что представленные в гимне понятия содержат "национальную смерть и катастрофу" для граждан.
The recent launch of OpenAI’s AI chatbot ChatGPT has raised alarms within Google, according to reports from The New York Times. Now, the Times says Google has plans to “demonstrate a version of its search engine with chatbot features this year” and unveil more than 20 projects powered by artificial intelligence.
As recently as December, we’d heard Google execs were worried that despite investing heavily in AI technology, moving too fast to roll it out could harm the company’s reputation. But things are changing quickly. Earlier this morning, Google announced it’s laying off more than 12,000 employees and focusing on AI as a domain of primary importance.
There’s no specific timeframe mentioned for the launch of Google’s AI search demo, but other projects from the slide presentation viewed by the Times are apparently set to debut during its annual I/O event in May, which has previously launched features like Duplex and Google Glass.
The situation is serious enough that the Times also reports that Google founders Larry Page and Sergey Brin — who remain the controlling shareholders of Google’s parent company Alphabet — have “offered advice” to company leaders, approved plans, and pitched ideas while meeting with executives to talk about ChatGPT last month. It notes that this level of reengagement, a shift after they left their daily roles in 2019, came at the invitation of Google / Alphabet CEO Sundar Pichai.
The demo for the chatbot search says Google will prioritize “getting facts right, ensuring safety and getting rid of misinformation,” hoping to address the issue of AI responding to queries confidently and clearly with bad information. Meanwhile, it’s also working on ways to speed up review processes that are supposed to check the technology to see if it’s operating in ways that are fair and ethical.
New product launches mentioned in the report, presented by a panel of execs including Jeff Dean, who runs its research and AI department, include an image generation studio that “creates and edits images,” an app for testing product prototypes, and a set of tools that other businesses can use to create AI prototypes from within a browser window called MakerSuite. The company is also working on a code generation tool called PaLM-Coder 2 that’s similar to Microsoft’s GitHub Copilot software and another that helps build apps for smartphones named Colab + Android Studio.
In recent years, Google has trodden carefully when it comes to the release of new AI products. The company found itself at the center of a debate over the ethics of artificial intelligence after firing two prominent researchers in the field, Timnit Gebru and Margaret Mitchell. The pair laid out criticisms of AI language models, noting challenges like their propensity to amplify biases in their training data and present false information as fact.
Although Google’s AI research is thought to be as advanced as that of other prominent tech companies, it’s only tested software with particularly restrictive guardrails. The firm’s AI Test Kitchen app, for example, offers access to image and text generation tools similar to OpenAI’s DALL-E and ChatGPT. However, Google heavily restricts the requests users can make of these systems. The company already showed off some of its own chat-heavy AI products, including a nonpublic demo in 2021 of a system similar to ChatGPT.
With the launch of OpenAI’s ChatGPT, though, and alarmist warnings about Google’s impending demise, it seems the company is revising its tactics. In the past, Google has said it’s avoided launching certain AI products because of the potential “reputational damage.” Now, it seems the reputation it wants to avoid is one of being behind the times.
Взято отсюдаAlongside eight states, the US Department of Justice is suing Google to break up the company’s advertising business. In a complaint filed Tuesday with a federal court in Virginia, the agency accused Google of illegally monopolizing the digital advertising market. “Google’s anticompetitive behavior has raised barriers to entry to artificially high levels, forced key competitors to abandon the market for ad tech tools, dissuaded potential competitors from joining the market, and left Google’s few remaining competitors marginalized and unfairly disadvantaged,” the Justice Department alleges.
"Today’s lawsuit from the DOJ attempts to pick winners and losers in the highly competitive advertising technology sector,” a Google spokesperson told Engadget. “It largely duplicates an unfounded lawsuit by the Texas Attorney General, much of which was recently dismissed by a federal court. [The] DOJ is doubling down on a flawed argument that would slow innovation, raise advertising fees, and make it harder for thousands of small businesses and publishers to grow."
As Bloomberg notes, the lawsuit represents the Biden administration’s first significant attempt to challenge the power of one of the nation’s largest tech firms. The agency previously sued Google in 2020. At the time, the Justice Department, under Attorney General William Barr, said the company had a monopoly over search and search-related advertising. It also took issue with the terms around Android, which the Justice Department said unfairly advantage Google by forcing manufacturers to preload their devices with the company’s applications and search engine.
Google faces intense regulatory scrutiny over its hold on the digital advertising market. In 2020, Texas filed a multi-state lawsuit accusing the company of using its "monopolistic power to control" ad pricing. One year later, the European Commission opened a probe into the company’s advertising business, a move that seems to have forced Google to reconsider how it handles ads on YouTube. Last year, the Senate also introduced legislation designed to prevent companies like Google from participating in more than one part of the digital advertising ecosystem.
“Having inserted itself into all aspects of the digital advertising marketplace, Google has used anticompetitive, exclusionary, and unlawful means to eliminate or severely diminish any threat to its dominance over digital advertising technologies,” the Justice Department says in its most recent complaint. It accuses Google of using acquisitions to take out both “actual or potential” competitors, in addition to abusing its marketplace dominance to prevent publishers and advertisers from using competing products effectively. “Whenever Google’s customers and competitors responded with innovation that threatened Google’s stranglehold over any one of these ad tech tools, Google’s anticompetitive response has been swift and effective,” the Justice Department alleges.
In a blog post published after news of the Justice Department’s lawsuit broke, Google accused the agency of “attempting to rewrite history at the expense of publishers, advertisers and internet users.” Specifically, Google says it takes issue with the Justice Department’s demand that the company spin off AdMeld and DoubleCheck, two ad tech firms Google acquired more than a decade ago. “These deals were reviewed by regulators, including by [the] DOJ, and allowed to proceed,” Google says.
Additionally, Google contends competition in the ad tech sector has increased significantly in recent years. For example, the company points to Microsoft’s recent purchase of AT&T’s former ad tech wing Xandr and notes the acquisition enabled Microsoft’s “landmark” advertising deal with Netflix. “The government did not challenge this acquisition,” Google points out.
Google also accuses the Justice Department of mischaracterizing how its advertising products work. The company says its product stack works with competing technologies, making it “easy” for publishers and advertisers to choose the services they want to use. “No one is forced to use our advertising technologies – they choose to use them because they’re effective,” Google contends.
One estimate suggests Google controls as much as 26.5 percent of the US digital ads market. The company’s ad unit is expected to generate about $73.8 billion in US ad revenue over the next year, with much of that money coming from search advertisements.
Update 4:04PM ET: Added more context from Google.
Взято отсюда
Google’s decision to let go of 12,000 employees was only just announced on Friday, but it extended the recent trend of the “Big Tech” companies cutting jobs in previously unheard-of numbers, and now we’ve seen more reports about where those cuts happened.
The Information reports that at Google, layoffs spread through nearly every group, including projects like Chrome, Search, Android, and Google Cloud. Its sources said they affected people who’d previously received “high performance reviews” and some managers making anywhere from $500K to $1 million.
One area of the search and advertising giant that was “relatively unaffected,” however, was the Google Brain division run by Jeff Dean, the senior vice president of research and artificial intelligence. That’s the team developing the machine learning tech Google is already using in many areas. According to the New York Times, its work will also be applied to a number of products we’ll reportedly see at its I/O event in May, including tools to generate images, a YouTube green screen feature, and at some point this year, a chatbot version of its search engine.
Google confirmed to Bloomberg that it decided “to wind down the majority of the Area 120 team.” That incubator, named for Google’s famed but long-dormant policy of allowing employees 20 percent of their time to chase side projects, had already seen large cuts affecting half of its projects late last year. The company said three projects will be folded into Google while the rest appear to be gone.
At Microsoft, where 10,000 people are being laid off, Polygon points out reports of the effects on several game development studios and the responses from some former employees. According to Bloomberg, there were cuts at Starfield developer Bethesda, and reporter Jason Schreier said in a tweet that its Halo studio 343 Industries was reportedly “hit hard.” At the same time, Kotaku noted that The Coalition, which works on the Gears of War games, was also affected.
Other cuts hit Microsoft’s mixed-reality teams. A report from Bloomberg indicates that layoffs affected some of the employees working on its HoloLens headset, a potential side effect of Congress denying the US Army’s request to purchase up to 6,900 headsets based on the HoloLens platform. Microsoft’s also shuttering the AltspaceVR virtual reality-based social platform it acquired in 2017, and employees on Microsoft’s Mixed Reality Toolkit team say they’ve all been laid off.
Amazon confirmed to us months ago that its layoffs included jobs in the devices and services division. Then, in early January, hardware chief Dave Limp said on CNBC’s TechCheck that the cuts affected nearly 2,000 people in his division, which is home to products like Alexa and its Echo smart home devices. Projects confirmed dropped included its video-calling device for kids and telehealth service.
This week, CNBC reports the layoffs also include a “significant number” of employees working on the Prime Air drone delivery project at multiple sites, including its headquarters, and, according to a since-deleted LinkedIn post, as much as half the team at its Pendleton, Oregon test site. Originally announced by Jeff Bezos in 2013, the program is seeing those cuts just as it starts to roll out testing in a few cities and prepares to launch the next-generation MK30 drone that will follow up the earlier MK27-2 unit shown above.
Amazon declined to say to CNBC how many people in the division had been laid off.
Update, 12:30PM ET: Updated to note the shutdown of AltspaceVR, as well as other layoffs in Microsoft’s mixed-reality division.
Взято отсюдаAfter steadily increasing prices over the past couple of years, Tesla has cut them drastically across its lineup in the US and Europe, in an apparent bid to boost sales. The least expensive EV, the Model 3 RWD, has dropped from $46,990 to $43,990, while the 5-seat Model Y Long Range fell 20 percent from $65,990 to $52,990. That means the latter model now qualifies for the $7,500 US Federal Tax credit, so the final price drop will be $20,500 — over 30 percent.
All other models, including the Model S and Model X (both regular and plaid versions) and the performance versions of the Model 3 and Model Y saw price cuts. However, the 7-seat option on the Model Y has increased in price by $1,000 to $4,000. While that puts the price over $55,000, it still qualifies for the Tax Credit as it's considered to be an SUV. In Germany, Model 3 and Y prices were cut by one to 17 percent, and also dropped in Austria, Switzerland and France.
To receive the full $7,500 rebate, customers will have to order before mid-March, when it's expected to be reduced to half that. The tax rebates apply to SUVs priced below $80,000 and cars under $55,000. The Model Y fell into both categories, depending on the number of seats, meaning certain 5-seat models weren't eligible — something CEO Elon Musk called "messed up." Now, all Model Y EVs except the Performance version appear to qualify.
Late last year, it was reported that Tesla inventory was building up after years of strong sales, first with reports in China, then the US. Tesla recently cut prices in China, then later declined any compensation to customers who complained. The Model 3 now starts at 229,900 yuan ($33,425) in that nation.
Neither Tesla nor CEO Elon Musk have weighed in on the price cuts, but potential buyers will no doubt be pleased (and recent purchasers displeased). The Model 3 was originally supposed to cost $35,000 and shipped in that configuration for a short time, but saw steady increases to the $46,990 price announced in March 2022. Last month, Musk said that "radical interest rate changes" had affected vehicle affordability and that Tesla could potentially lower prices.
Взято отсюдаThe premise seems simple, according to Amtrak’s website: “Instead of being stuck in traffic on I-95, you and your vehicle could Get Carried Away on the Auto Train.”
Pack your car with everything you need for a vacation, drive to the Auto Train station in Lorton, Virginia (near Washington, D.C.), for an overnight trip to Sanford, Florida (near Orlando), and presto — you save 900 miles of driving during a 17 1/2-hour trip.
It sounds fantastic … until this week’s Auto Train nightmare.
Passengers who departed Lorton on Monday at 5 p.m. didn’t arrive in Sanford until Wednesday at 6 a.m. — a 20-hour delay.
In total, they spent 37 hours stuck on board, mostly in South Carolina, after an earlier derailment elsewhere on the line.
Reports of people and pets unable to disembark, passengers calling 911 to say they were being held hostage, and a single delivery of pizza during the entire two-day delay have rocketed through the news.
Here’s what we know so far about the incident, as well as information about train passenger rights.
Amtrak’s overnight Auto Train was on its way from Virginia to Florida. However, according to Kimberly Woods, Amtrak’s senior public relations manager: “The Amtrak Auto Train had significant delays due to a CSX freight derailment in South Carolina. The train was detoured off its normal route in order to continue operating south, adding additional time to the trip.”
Although news outlets have reported many passenger complaints that include a lack of communication and access to outdoor space, Amtrak maintains that “Customers were provided regular updates, along with meals, snack packs, and beverages. Additional food from local businesses along the route was also provided during the trip. The onboard staff worked with pet owners to provide bathroom breaks.”
We hate to be the bearer of bad news, but as you may have already surmised, there are no passenger protections for rail travel in the U.S.
Much like Europe’s EU261 flight compensation laws for air travel, most European countries offer immediate compensation for rail delays. For example, in Germany, delays of just 60 minutes start the monetary compensation process.
The U.S. Department of Transportation very recently instituted new policies for U.S. airlines to make it easier for passengers to receive refunds for delayed flights. However, the department has not yet taken similar actions for rail travel.
It is up to Amtrak to work to offer restitution to passengers involved in this week’s train mishap. We reached out to the company for details on how it plans to compensate the passengers involved but have not heard back as of midday Wednesday.
There should be a simple answer for who regulates Amtrak, but it’s a little cloudy.
In 2015, in a Supreme Court case involving Amtrak and freight trains, Justice Anthony Kennedy wrote in the majority opinion, “Amtrak was created by the government, is controlled by the government, and operates for the government’s benefit.”
Additionally, the DOT oversees passenger rail service. However, according to the DOT, Amtrak is run as a for-profit company with the U.S. Secretary of Transportation (currently Pete Buttigieg) holding a seat on the 10-person board.
Взято отсюда
a_nikolov
