Please try the URL privacy information feature enabled by clicking the flashlight icon above. This will reveal two icons after each link the body of the digest. The shield takes you to a breakdown of Terms of Service for the site - however only a small number of sites are covered at the moment. The flashlight take you to an analysis of the various trackers etc. that the linked site delivers. Please let the website maintainer know if you find this useful or not. As a RISKS reader, you will probably not be surprised by what is revealed…
https://www.theverge.com/2024/4/8/24124189/solar-eclipse-renewable-energy-panels-electricity-grid [And incidentally, Monty noted Internet Traffic Dipped as Viewers Took in the Eclipse Internet—it dropped by 40 percent or more during the eclipse in states in the path of totality, including Maine, New Hampshire and Ohio, Cloudflare found. https://www.nytimes.com/2024/04/09/business/internet-traffic-eclipse-cloudflare.html PGN]
Kim S. Nash. *The Wall Street Journal* https://cybersecurity.cmail20.com/t/d-l-eydzx-tjludishy-i/ Security missteps at Microsoft might pave the way for a cyber-overhaul of the cloud sector. "A cascade of security failures at Microsoft" allowed Chinese hackers to penetrate the company's Exchange Online cloud-based email system last year, according to the U.S. Cyber Safety Review Board. The board spent seven months investigating the espionage incident, in which the email accounts of 22 organizations and more than 500 individuals around the world were compromised. These included Commerce Secretary Gina Raimondo and U.S. Ambassador to China, Nicholas Burns. Microsoft cooperated fully with the probe, the board said in its report issued Tuesday. Tactical and strategic decisions at Microsoft reflect "a corporate culture that deprioritized enterprise security investments and rigorous risk management, at odds with the company's centrality in the technology ecosystem and the level of trust customers place in the company to protect their data and operations," the board said. Not only does the report include security recommendations for the tech giant, but some for cloud providers generally as well. U.S. officials also plan to convene major cloud players to hammer out baseline cyber practices and a process for the companies to regularly attest they are complying.
The episode has imperiled the reputation of the Harvard-affiliated Dana-Farber Cancer Institute and raised questions about the work of one high-profile researcher. https://www.nbcnews.com/science/science-news/cancer-institute-dana-farber-retracts-studies-errors-rcna143922
Artificial intelligence tools are only making it easier to spot problems. Some scientists say it’s time for universities and academic publishers to reform how they address flawed research. https://www.nbcnews.com/science/science-news/-ignored-community-science-sleuths-now-research-community-heels-rcna136946
Providers of mental health services are turning to AI-powered chatbots
designed to help fill the gaps amid a shortage of therapists and growing
demand from patients.
But not all chatbots are equal: some can offer helpful advice while others
can be ineffective, or even potentially harmful. Woebot Health uses AI to
power its mental health chatbot, called Woebot. The challenge is to protect
people from harmful advice while safely harnessing the power of artificial
intelligence. [...]
The National Eating Disorders Association's AI-powered chatbot, Tessa, was
taken down after it provided potentially harmful advice to people seeking
help.
https://www.cbsnews.com/news/ai-chatbots-mental-health-therapy-pitfalls-60-minutes/
They're so proud—rules-based bot gives deterministic response. Same
input, same output. Guidelines/guardrails protect against anything bad --
except when they're modified and don't. GPT bot improvises.
What could go wrong?
[Well, they might need a Woebot Wabbit? PGN]
https://techcrunch.com/2024/04/08/hackers-stole-340000-social-security-numbers-from-government-consulting-firm/
https://arstechnica.com/?p=2015480
https://techcrunch.com/2024/04/08/targus-says-cyberattack-causing-operational-outage/
https://arstechnica.com/?p=2015412
Keaton Peters, *The Texas Tribune*, 9 Apr 2024 The Texas Education Agency (TEA) this year will use an "automated scoring engine" that uses natural language processing technology to assess and grade open-ended questions on the State of Texas Assessment of Academic Readiness (STAAR) for reading, writing, science, and social studies. TEA gathered 3,000 responses that went through two rounds of human scoring, and used them to teach the automated scoring engine the characteristics of responses. It is programmed to assign the same scores a human would have given. [Texas is leaning to STAARBOARD for a change? But this loo ks more like any PORT in a storm. What was the quality of the 3,000 human scorings? The system could have been trained on sloppy grading, and assigning the same scores may be really wrong-headed. Once again, we desperately need sound evidence-based results. PGN]
"If you don't know where you're going, any road will take you there." -- Cheshire Cat in Alice's Wonderland Isn't this jamming/spoofing a fraud on the location-based advertisers ? How soon before we have GPS 'swatting' on Carmen Sandiego ? https://en.wikipedia.org/wiki/Swatting https://www.newscientist.com/article/2415318-ukraine-will-spoof-gps-across-the-country-to-stop-russian-drones/ Ukraine will spoof GPS across the country to stop Russian drones
Many job seekers often change their LinkedIn status to #OpenToWork. Scammers look for people who do that and launch scams against them. Most often around resume building, executive coaching, and job introductions. https://brothke.medium.com/when-opentowork-is-really-opentoscam-598ef27dd628?sk=b65fb880100304aa67a53a0590c7b162
The Horizon post office scandal in the UK has been discussed periodically in RISKS over the years. The docudrama that caused the UK parliament to finally take action is now available in the US on PBS. https://www.pbs.org/wgbh/masterpiece/shows/mr-bates-vs-the-post-office/ I'm curious to see whether viewers will see the parallel to trusting AI systems, and the risks those bring.
How The Commons Was Closed for the Benefit of the Few https://danafblankenhorn.substack.com/p/why-open-source-cant-innovate
Musk is being sued for falsely suggesting a 22-year-old Jewish man was part of a neo-Nazi brawl. The lawsuit against the billionaire, filed in October, alleges that Musk used his colossal social media platform to amplify a false far-right conspiracy theory linking 22-year-old Ben Brody to a brawl in Oregon between the neo-Nazi group Rose City Nationalists and the Proud Boys, a neo-fascist fight club. The brawl occurred during Oregon City’s first Pride Night Fest, when both groups came to disrupt the event and spew anti-LGBTQ+ rhetoric. Brody wasn't even in the same state when the June 24 brawl occurred. But his world was turned upside down when far-right X accounts, magnified by Musk, falsely identified him as a member of Rose City Nationalists (and an undercover federal agent) and posted his personal information online. Musk amplified the conspiracy theory repeatedly to his more than 180 million followers, suggesting Brody was a fresh-faced federal agent pretending to be a neo-Nazi in a “false flag situation,” a phrase used to suggest a harmful event was deliberately set up to misrepresent a group or person. [...] Brody said he and his family were forced to flee their home amid the fallout from Musk’s posts. He’s seeking more than $1 million in damages. The next court hearing is scheduled for April 22. [...] “People are attacked all the time in the media, online media, social media, but it is rare that that actually has a meaningful negative impact on their life,” Musk testified. https://www.huffpost.com/entry/elon-musk-didnt-want-his-latest-deposition-released-here-it-is_n_66133d2ce4b0d81853f9a766
In a campaign stoking anti-Ukraine sentiment in the U.S., Russia-directed trolls have written thousands of fabricated news articles and social media posts. https://www.washingtonpost.com/world/2024/04/08/russia-propaganda-us-ukraine/
Stewart Lucas Murrey is suing more than 50 women for sharing stories about him in multiple private Facebook groups. https://www.nbcnews.com/tech/judge-dismisses-are-we-dating-same-guy-facebook-group-lawsuit-rcna147043
This is the most consequential technology in America This is America’s most popular social app by a mile, the top way to listen to music, the healthiest economy on the Internet and essential AI training fuel. (Spoiler alert: It’s YouTube.) You think you know YouTube. It’s where billions of people learn how to change a tire, follow a favorite yoga workout or catch footage of Monday’s solar eclipse. But maybe you don’t know that YouTube is also the most popular way to hear music and one of the country’s largest cable TV providers. YouTube is the healthiest economy on the Internet. And it has been rocket fuel for artificial intelligence. I’m digging into YouTube’s identity because it’s essential to understand the influence of technologies in our lives. As popular as YouTube is, its power over the Internet and us is somehow still underrated. Let me try to persuade you that YouTube is the most consequential technology in America: [...] https://www.washingtonpost.com/technology/2024/04/09/most-important-app-youtube/
So, I got this email inviting me to a Trello workspace. I assume Trello is something like Slack or MS Teams. I initially assumed that this was yet another instance of someone assuming that "rslade@gmail.com" was *their* email address, rather than mine. But then I saw the included note: "Greetings, I am Brian Smith, a seasoned attorney at Piccadilly's Attorneys Firm and Personal attorney to a deceased member of your family who died and left behind an Estate claim. You have been designated as a devisee. Please contact me via email for further information. [attorneybriansmith96@gmail.com]" OK, right, it's just another version of the "your rich relative died" advance fee fraud. But I found it interesting that they are trying yet another workaround to get by standard spam filters ...
https://www.theverge.com/2024/4/8/24124744/tesla-autopilot-lawsuit-settlement-huang-death
> Actually, using face-recognition methods may be the most humane way to > tell apart terrorists who hide among the civilian population. Especially > when the alternative older methods were more like "kill them all and let > God sort them out". Facial recognition should be forbidden from use by law enforcement unless and until it is able to be used on white collar criminals (tax evasion, securities fraud, insider trading, etc.). The actual losses from those crimes dwarfs losses from all other types of crime facial recognition will pitched as the solution for.
Please report problems with the web pages to the maintainer