Microsoft AI ethics department disbanded amidst industry warnings

Microsoft has laid off its entire Ethics & Society team, a group responsible for ensuring that product designs reflect the company’s principles pertaining to artificial intelligence development. The move, which came as a surprise, has raised concerns among experts, especially considering that Microsoft has invested over $11 billion in OpenAI, which powers Microsoft’s chat program, Bing, and other services.

Microsoft has made the decision as part of a series of dramatic layoffs, affecting 10,000 employees or roughly 5 percent of the company’s global workforce. The cutbacks were made at a time when Microsoft, Google, and Meta were embarking on a new “AI arms race” to offer consumers chat programs, text generators, and revolutionary online search tools. Although Microsoft has an Office of Responsible AI dedicated to overseeing artificial intelligence initiatives, there is a disconnect between how those principles and guidelines are implemented in the actual projects.

Critics of Microsoft’s move argue that having a dedicated ethics team is crucial to ensure responsible AI development. Duri Long, an assistant professor at Northwestern University, says, “This technology is new, and we are still learning about its implications for society. Dedicated ethics teams are vital to the responsible development of any technology, and especially so with AI.” Although Microsoft’s corporate VP of AI has said that the team’s responsibilities are being devolved to product teams, concerns remain over the lack of oversight and the need for transparency in the development of AI technology.

In conclusion, Microsoft’s decision to dissolve its Ethics & Society team has raised concerns about the responsible development of AI technology. In the midst of the development of new AI products, companies and others in the tech industry are urging caution against unleashing software that is hastily tested and frequently problematic. It remains to be seen how Microsoft will address the gap between its Office of Responsible AI and the company’s AI projects and ensure that ethical considerations are taken into account in its AI development in the future.

Read: https://www.popsci.com/technology/microsoft-ai-team-layoffs/


FBI Analyst violates rules in Congress member search, reignites section 702 debate

An FBI analyst has been found to have violated rules by conducting excessively broad searches regarding an unnamed member of Congress during 2019 or 2020. The discovery of this violation was disclosed in a report that was declassified in December.

During a hearing before the House Intelligence Committee, Representative Darin LaHood publicly identified himself as the member of Congress who was inappropriately queried on multiple occasions. LaHood leads a group of Intelligence Committee members from both parties working to persuade Congress to renew the warrantless surveillance law, Section 702, which has come under scrutiny due to privacy and civil liberties concerns.

Despite the violation, LaHood believes that Section 702 is important for combatting foreign threats and that Congress must renew the law. However, civil libertarians and some Republicans who share former President Trump’s distrust of security agencies and surveillance are likely to challenge the law’s reauthorization.

LaHood has criticized the searches as an egregious violation that could be seen as a threat to the separation of powers. He emphasized that the government needs to do more to build trust with the public and Congress to secure the extension of the law.

Section 702 of the Foreign Intelligence Surveillance Act (FISA) allows the government to collect information on non-U.S. citizens who are located abroad. Still, it also permits incidental information collection on U.S. citizens who may communicate with those foreign targets.

Critics argue that this incidental collection violates Fourth Amendment protections against unreasonable searches and seizures and that the law allows for the collection of vast amounts of data without sufficient oversight or transparency.

The issue of warrantless surveillance is contentious, with competing interests of national security and civil liberties at stake. The revelation of the violation by the FBI analyst raises concerns about the government’s ability to use the surveillance powers granted by Section 702 responsibly.

Read: https://www.nytimes.com/2023/03/09/us/politics/fbi-surveillance-darin-lahood.html

Jonathan Turley: Senator Kelly Adds Bank Insolvency to the Long List of Subjects Democrats Wish to Censor on Social Media

In his new column over at the New York Post, Jonathan Turley speaks about the latest example of Democrats’ supposed attempt at social media censorship. Citing the example of Sen. Mark Kelly (D., Ariz.) who he says called for the censorship of bank critics, Turley warns of the increased frequency with which certain views are labelled as being simply too dangerous for social media.

In a Zoom call this week, Kelly asked representatives from the Federal Reserve, Treasury Department, and Federal Deposit Insurance Corporation to censor social media content to remove rising doubts over bank solvency in the wake of the Silicon Valley Bank and Signature Bank crises. Rep. Thomas Massie (R-Ky.) confirmed that Kelly suggested the “government should work with social media companies to censor information that could lead to a run on banks.” Kelly is further reported to have quoted the dangers of “foreign actors” that may be influencing Americans on social media to undermine US banks.

Turley says he sees bank solvency as yet another topic in the long list of subjects censored by Democrats, ranging from climate change and gender identity to how to deal with COVID-19 and the US federal elections. He is concerned by the support of Sen. Richard Blumenthal (D-Conn.), who warned that he and his colleagues would not tolerate any “backsliding or retrenching” by firms “failing to take action against dangerous disinformation.” Turley argues that the Democrats should be working to convince citizens that their deposits are safe instead of silencing the opposition.

He sums up his column by quoting the civil libertarian Justice Louis Brandeis, who said, “The greatest dangers to liberty lurk in insidious encroachment by men of zeal, well-meaning but without understanding.”

Read: https://www.zerohedge.com/political/banking-censorship-sen-kelly-becomes-latest-dem-suggest-barring-opposing-views-social


Cybercrime Forum BreachForums’ Administrator Arrested in New York

The FBI arrested Conor Brian Fitzpatrick, the alleged administrator of BreachForums, a well-known cybercrime forum where some of the world’s largest hacked databases are sold. Fitzpatrick, who used the alias “pompompurin,” had been a thorn in the FBI’s side for several years. BreachForums is widely believed to be a reincarnation of RaidForums, a similar crime forum that the FBI infiltrated and dismantled in 2022.

Fitzpatrick has been accused of conspiracy to commit access device fraud and was arrested on March 15, 2023, in Peekskill, NY. He claimed responsibility for a 2021 email blast sent from FBI email systems and internet addresses. According to him, he was able to exploit a flaw in an FBI portal to send fake emails. The FBI later acknowledged a software misconfiguration allowed someone to send the fake emails.

BreachForums was the sales forum for data stolen from DC Health Link, a health insurance exchange in Washington, D.C., that recently suffered a data breach. While the official notice about the breach states 56,415 people were affected, the sales thread claimed 170,000 people’s names, Social Security numbers, dates of birth, and health plan information was compromised. BreachForums members are now concerned that the FBI may have access to their registration details.

Despite Fitzpatrick’s arrest, BreachForums remains accessible online, and active users on the site are only just discovering that the FBI likely has access to their administrator and the site’s database.

Read: https://krebsonsecurity.com/2023/03/feds-charge-ny-man-as-breachforums-boss-pompompurin/


Digital Surveillance on the rise in Australian workplaces, with Commonwealth Bank in the spotlight

Commonwealth Bank (CBA) uses an app called ‘Navigate’ to monitor its employees’ online activity and computer usage. While the revelation has led to anger among employees, it turns out that 90% of Australian employers use tracking software on their computers and phones. The pandemic has led to more companies using technology for employee surveillance. Almost every Australian company uses monitoring software on its computers and phones, compared with the global average of 80%, according to Herbert Smith Freehills.

Tracking keyboard use, typing, mouse movements, and location is marketed as a measure of productivity and efficiency, but Professor Peter Holland of Swinburne University in Melbourne questioned its purpose. He said that companies should consider what they are trying to achieve and if they have discussed this with their workforce.

The Finance Sector Union (FSU) has questioned the bank’s use of the Navigate app, and its national secretary Julia Angrisano has criticized the bank for not notifying the union that it has instigated widespread employee surveillance. According to reports, the bank also uses office attendance data and computer software to monitor absenteeism, including early marks and long lunches, and has ordered some employees to take leave if they aren’t performing well.

In her remarks, Ms. Angrisano said workers had a right to know what information was being collected and that monitoring employees shouldn’t be used for intimidation or pressure.

In Australia, employees are heavily monitored, with some receiving emails after leaving their desks for a short time. The advice to workers is to use private devices for private matters. But even if using personal devices, such as phones and computers, employees can still be tracked if logged into company software. The high level of surveillance in Australia means that employees may be being watched without their knowledge.

Read: https://www.dailymail.co.uk/news/article-11865259/Commonwealth-Bank-isnt-company-track-workers-computers-phones.html

The Virality Project

“Researchers at Stanford University—in partnership with several nonprofits that have received government funding—worked with social media platforms to flag and suppress commentary on COVID vaccines, science, and policy that contradicted public health officials’ stances, even when that commentary was true.”

Read: https://reason.com/2023/03/17/researchers-pressured-twitter-to-treat-covid-19-facts-as-misinformation/


Elsewhere online:

Pig butchering crypto scams: increasing sophistication and psychological manipulation target victims
Read: https://cyware.com/news/fbi-warns-users-about-surge-in-pig-butchering-crypto-scams-ae798a85/

Marvel Seeks Reddit’s Help to Unmask Quantumania Script Leakers
Read: https://www.avclub.com/marvel-studios-wants-reddit-unmask-quanumania-leakers-1850222399

Russia-linked APT29 Spotted Abusing Legitimate EU Information Exchange Systems in New Cyber Espionage Campaign
Read: https://securityaffairs.com/143545/apt/apt29-abused-information-exchange-systems.html

CISA Analysts Report Multiple Exploitations of Progress Telerik .NET Deserialization Vulnerability
Read: https://www.cisa.gov/news-events/alerts/2023/03/15/threat-actors-exploited-progress-telerik-vulnerability-us-government-iis-server

American Members of Notorious ViLE Cybercrime Group Charged for Hacking into the US DEA Portal in 2022
Read: https://www.hackread.com/us-citizens-charged-hacking-dea/