Federal agencies are the latest to alert companies to potential bias in AI recruiting tools. As the APnotes, the Justice Department and Equal Employment Opportunity Commission (EEOC) have warned employers that AI hiring and productivity systems can violate the Americans with Disabilities Act. These technologies might discriminate against people with disabilities by unfairly ruling out job candidates, applying incorrect performance monitoring, asking for illegal sensitive info or limiting pay raises and promotions.

Accordingly, the government bodies have released documents (DOJ, EEOC) outlining the ADA's requirements and offering help to improve the fairness of workplace AI systems. Businesses should ensure their AI allows for reasonable accommodations.They should also consider how any of their automated tools might affect people with various disabilities.

There's no guarantee companies will follow the advice. However, it comes amid mounting pressure on companies to temper their uses of AI for recruiting and worker tracking. California recently enacted a productivity quota law banning algorithms that violate health, labor and safety regulations, or lead to firings of people who can't meet dangerous quotas. New York City, meanwhile, now requires that AI hiring systems pass yearly audits looking for discrimination. Companies that don't heed the new warnings could face serious legal repercussions at multiple levels.

US Senator Michael Bennet (D-CO) wants to establish a federal watchdog that would focus on overseeing digital platforms and tech giants. The lawmaker has introduced the Digital Platform Commission Act (PDF) in Congress in hopes of establishing a five-person federal body appointed by the President and approved by the Senate. They would be experts in relevant fields, including computer science, software development and technology policy.

The commission would be in charge of assuring "the fairness and safety of algorithms on digital platforms" as well as promoting competition. It would also have the authority to conduct investigations, impose penalties and to set new rules, such as those that ensure moderation transparency and the protection of consumers. The commission would create requirements for regular public risk assessments on the distribution of harmful content on digital platforms, as well.

Under the commission, a "Code Council" comprised of technologists and public interest experts will conjure up standards and policies that could be implemented. In addition, the commission will establish a research office with 20 dedicated employees to conduct internal research and coordinate with outside academics and experts. 

As mentioned in the legislation's announcement, the Department of Justice and the Federal Trade Commission are in charge of overseeing digital platforms today. Bennett argues, however, that they lack the expert staff and tech-oriented culture necessary for robust oversight. 

The Washington Post reports that Bennett's motivation was his personal experience viewing disinformation as part of the Senate Intelligence Committee, as well as seeing how social media has affected his children. As the publication notes, though, it remains to be seen whether the legislation would be approved by the Senate, where Democrats have a 50-50 majority.

The 5th US Circuit Court of Appeals has paused the temporary injunction on controversial law HB 20, which another court blocked from taking effect last year. As Houston Public Media notes, the state introduced HB 20 last year after high-profile conservatives, including Donald Trump, were blocked on social media websites. Under the law, users will be able to sue large social media platforms with more than 50 million active monthly users such as Facebook and Twitter if they believe they were banned for their political views. HB 20 also prohibits social networks from removing or restricting content based on "the viewpoint of the user or another person." 

Trade industry groups NetChoice and the Computer and Communications Industry Association (CCIA) managed to secure an injunction against the law last year. They argued that HB 20 would lead to the spread of misinformation and hate speech on social networks and that it also violates the websites' First Amendment rights. The federal judge overseeing the case agreed that social networks have the right to moderate content under the First Amendment and also said that parts of the law are "prohibitively vague."

In a hearing for the appeal filed by Texas, the state's lawyers argued that social media platforms are "modern-day public squares." That means they can be required to host content that they deem objectionable and are banned from censoring certain viewpoints. The 5th Circuit judges sided with Texas, with one even telling the trade groups during the hearing that social networks like Twitter are not websites but "internet providers" instead.

NetChoice counsel Chris Marchese called HB 20 "an assault on the First Amendment" and "constitutionally rotten from top to bottom" on Twitter. The trade groups plan to appeal immediately, but for now, HB 20 is fully in effect. 

A federal court blocked a similar law in Florida last year after the judge ruled that it violates Section 230 of the Communications Decency Act that shields online platforms from liability for what their users' post. Florida also appealed that decision, which will be decided by the 11th Circuit Court of Appeals.

Although it's supposed to be restricted by surveillance rules at local, state and federal levels, Immigration and Customs Enforcement (ICE) has built up a mass surveillance system that includes details on almost all US residents, according to a report from a major think tank. Researchers from Georgetown Law's Center on Privacy and Technology said ICE "now operates as a domestic surveillance agency" and that it was able to bypass regulations in part by purchasing databases from private companies. 

"Since its founding in 2003, ICE has not only been building its own capacity to use surveillance to carry out deportations but has also played a key role in the federal government’s larger push to amass as much information as possible about all of our lives," the report's authors state. "By reaching into the digital records of state and local governments and buying databases with billions of data points from private companies, ICE has created a surveillance infrastructure that enables it to pull detailed dossiers on nearly anyone, seemingly at any time."

The researchers spent two years looking into ICE to put together the extensive report, which is called "American Dragnet: Data-Driven Deportation in the 21st Century." They obtained information by filing hundreds of freedom of information requests and scouring more than 100,000 contracts and procurement records.

The agency is said to be using data from the Department of Motor Vehicles and utility companies, along with the likes of call records, child welfare records, phone location data, healthcare records and social media posts. ICE is now said to hold driver’s license data for 74 percent of adults and can track the movement of cars in cities that are home to 70 percent of the adult population in the US.

The study shows that ICE, which falls under the Department of Homeland Security, has already used facial recognition technology to search through driver’s license photos of a third of adults in the US. In 2020, the agency signed a deal with Clearview AI to use that company's controversial technology. In addition, the report states that when 74 percent of adults hook up gas, electricity, phone or internet utilities in a new residence, ICE was able to automatically find out their updated address.

The authors wrote that ICE is able to carry out these actions in secret and without warrants. Along with the data it acquired from other government departments, utilities, private companies and third-party data brokers, "the power of algorithmic tools for sorting, matching, searching and analysis has dramatically expanded the scope and regularity of ICE surveillance," the report states.

Spending transactions reviewed by the researchers showed that, between 2008 and last year, ICE spent around $2.8 billion on "new surveillance, data collection and data-sharing initiatives." It spent approximately $569 million on data analysis, including $186.6 million in contracts with Palantir Technologies to help it make sense of its vast troves of data. Records showed that ICE also spent more than $1.3 billion on geolocation tech during that timeframe and $389 million on telecom interception, which includes tech that helps the agency track someone's phone calls, emails, social media activity and real-time internet use.

In addition, the findings suggest the agency started engaging in certain surveillance activities much earlier than previously believed. The researchers found a contract from 2008 that granted ICE access to the Rhode Island motor vehicle department’s facial recognition database. Prior to that, it was understood that ICE started conducting facial recognition searches on state and local data sets in 2013.

The authors claim that ICE has been able to sidestep congressional oversight and bypass attempts at state level to curtain its surveillance capabilities. They included a list of recommendations that may help rein in the agency's surveillance dragnet, such as Congress reforming immigration laws to "undercut ICE surveillance authority" and blocking ICE's use of DMV data. The recommended measures also include protecting people who trust federal, state and local authorities with their data and blocking the use of utility records for immigration enforcement.

Engadget has contacted ICE and the Department of Homeland Security for comment.

Russia's alleged cyberattack campaign against Ukraine appears to have continued up to the very minutes before the invasion. Reutersreports the US, UK and European Union have formally blamed Russia for a large-scale cyberattack that disrupted Viasat's satellite internet service an hour ahead of the February 24th assault. The hacking efforts permanently destroyed "tens of thousands" of satellite terminals, Viasat said, while the UK noted that the attack affected central European internet users and wind farms in addition to the Ukranian military and some civilian customers.

The accusations come in response to "new UK and US intelligence" linking Russia to the cyberattack, according to the UK's Foreign Office. Russia hadn't responded to the claims as of this writing, but has historically denied cyberattacks regardless of evidence.

The hack targeting Viasat likely had its intended effect. Ukraine cybersecurity official Victor Zhora disclosed in March that the anti-satellite effort led to a "huge loss" in communications at the very start of the war. With that said, Ukraine has been better-prepared in at least some instances. It claims to have fended off an attack against an energy provider in April, for example. While this latest attribution won't dissuade Russia from conducting more attacks, it might signal that Ukraine and its allies are more aware of how to defend themselves online.

On the same day Russia celebrated its role in defeating Nazi Germany, many of the country’s online platforms were defaced in protest of the war in Ukraine. The Washington Post reported on Monday that Russians with smart TVs saw channel listings replaced with a message implicating them in the ongoing conflict. “The blood of thousands of Ukrainians and hundreds of murdered children is on your hands,” the message read, according to the outlet. “TV and authorities are lying. No to war.”

In addition to smart TVs, the apparent hack targetted some of the country’s largest internet companies, including Yandex. Hackers also went after Rutube, Russia’s alternative to YouTube. “Our video hosting has undergone a powerful cyberattack. At the moment, it is not possible to access the platform,” the service said in a statement it posted on its Telegram channel. Rutube later stated it had isolated the attack and that its content library wasn’t accessed in the incident.

Since Russia launched its invasion of Ukraine on February 24th, the country has come under consistent attack from hackers. In the early days of the conflict, Anonymous claimed responsibility for a series of DDoS attacks that left several official government websites, including one belonging to the Ministry of Defence, inaccessible. It’s believed Anonymous was also responsible for an incident that saw several Russian state TV channels play the Ukranian national anthem. At the same time, Ukraine, with help from Microsoft and other western companies, recently managed to prevent Russian military hackers from disrupting one of the country’s energy providers.

A key component of the Biden Administration's $1 trillion US Infrastructure Bill was ensuring that people of all means had access to reliable high-speed internet. Now, twenty internet providers including AT&T, Comcast and Verizon have agreed to offer high-speed broadband internet plans for no more than $30 per month via a subsidy, the White House announced.

The ISP's, which cover 80 percent of the US population, agreed to "either increase speeds or cut prices, making sure they all offer ACP-eligible households high-speed, high-quality internet plans for no more than $30/month," The White House wrote. Biden has previously highlighted the challenges not having broadband poses to some families. "Never again should a parent have to sit in their car in a McDonald’s parking lot... so that their child can get access to high-speed Internet to do their homework," he said in a speech last month. 

The subsidy is part of a $65 billion program for expanding broadband primarily through fiber-optic cable installations. $14 billion of that is earmarked for subsidies in the Affordable Connectivity Program aimed at lowering internet costs. It's available to any families with income 200 percent or less than federal poverty guidelines, or for those who qualify for certain assistance programs.

Some 11.5 million households have signed up for the subsidy, but there are as many as 48 million eligible households. To that end, the administration is launching a site called GetInternet.gov that will provide details on how to sign up. It's also reaching out to people through federal agencies, partnering with states and cities and collaborating with public interest organizations like the United Way and Goodwill. 

San Francisco federal district court Judge James Donato has tossed the lawsuit Donald Trump filed against Twitter last year in a bid to get his account back. The social network permanently suspended the former president's account after his supporters stormed the Capitol in January 2021. In the company's announcement, Twitter cited two of his tweets in particular that it believes were "highly likely to encourage and inspire people to replicate the criminal acts that took place at the US Capitol" on January 6th last year.

Trump filed a lawsuit in October, seeking a preliminary injunction on the ban and arguing that it violates his First Amendment rights. Donato disagreed and noted in his ruling that Twitter is a private company. "The First Amendment applies only to governmental abridgements of speech," he explained, "and not to alleged abridgements by private companies." The judge also rejected the notion that the social network had acted as a government entity after being pressured by Trump's opponents and had thereby violated the First Amendment when it banned the former President. 

In his lawsuit, Trump asked the judge to rule the federal Communications Decency Act, which states that online service providers such as Twitter can't be held liable for content posted by users, as unconstitutional. The judge shot down that claim, as well, and ruled that the former President didn't have legal standing to challenge Section 230 of CDA. Trump is a known critic of Section 230 and proposed to limit the protections social media platforms enjoy under it during his term.

The former President was an avid Twitter user before his suspension and formed his own social network called Truth Social after he was banned. Just recently, he told CNBC that he won't be going back to Twitter even if Elon Musk reverses his suspension and will stay on Truth Social instead. According to a recent report by the Daily Beast, Truth Social has 513,000 daily active users compared to Twitter's 217 million.

Cryptocurrency mixers are sometimes used to help online criminals launder their stolen money by hiding its true origins, and the US Treasury is now ready to clamp down on them when hostile governments are involved. The department has issued its first sanctions against a Bitcoin mixer, Blender.io, for allegedly and "indiscriminately" helping North Korea launder over $20.5 million in crypto from the $620 million Axie Infinity heist and other crimes. 

The measures block all Blender property in the US (or controlled by US residents), as well as US-linked transactions and any entities where blocked people have majority control. On a basic level, blocks create an audit trail and prevent sanctioned entities' funds from changing hands.

The sanctions come after officials pinned the Axie Infinity theft on Lazarus Group, an outfit frequently linked to the North Korean government's cybercrime and cyberwarfare efforts. North Korea has been repeatedly accused of hacking banks and cryptocurrency holders to evade international sanctions and finance its weapons programs.

The Treasury's Office of Foreign Assets Control also used the opportunity to identify four digital wallets Lazarus reportedly used to launder the rest of the Axie Infinity crypto. The perpetrators relied on one "getaway" wallet for the crime itself.

The agency stressed that most cryptocurrency activity was legal, and that it was only targeting mixers that aid criminals. However, there's a not-so-subtle warning here: the US is willing to sanction crypto service providers if they tolerate state-backed hackers, not just the nations directing those hacks. 

In a win for net neturality, ISPs agreed to end their legal challenge to a 2018 Californa law that bars providers from throttling service. Telecom groups and California Attorney General Rob Bonta today jointly agreed to dismiss the case, reportedReuters. 

It’s fair that say that luck hasn’t exactly been on the telecom industry’s side. Earlier this year, the 9th Circuit Court of Appeals refused to reconsider its ruling that California’s law be upheld. And last year, the US DOJ dropped its own lawsuit over the net neutrality law, which the agency had filed during the Trump administration.

“Following multiple defeats in court, internet service providers have finally abandoned an effort to block enforcement of CA's net neutrality law. This is a win for California and for a free and fair internet,” wrote Bonta in a tweet.

After Trump-appointed FCC Commissioner Ajit Pai overturned the agency's net neutrality rules in 2017, California’s legislature decided to enact its own law. The state’s net neutrality law, which went into effect in August 2018, expanded on previous federal rules by banning the use of “zero-rating” by ISPs in an anti-competitive manner. Zero-rating occurs when an ISP exempts any of its affiliated services from eating away at a customer’s data caps. For example, AT&T Wireless once exempted HBO Max from the data caps of its internet customers. The company dropped this practice last year, and blamed the impact of California’s law. Digital rights groups like Electronic Frontier Foundation have argued that zero-rating is hostile to consumers, especially those from low-income households.

Federal net neutrality rules that were blocked under the Trump administration have yet to be restored by the FCC under President Joe Biden. That’s because the five-member panel is currently short one member, which they’ll need in order to vote on net neutrality. The agency is awaiting the Senate confirmation of Gigi Sohn. But thanks to intense lobbying from telecom groups and a number of Republicans (and moderate Democrats) in Congress, Sohn’s confirmation is stalled at present.