OpenAI hits back at Elon Musk, says 90 percent of statistics published on the internet are not true, 30,000 US Fidelity customers had their data stolen and hospital alarms may not save lives Yes, they are the cause of many deaths.
All this and more in Hashtag Trends' “Lies, Bad Liars, and Internet Statistics'' edition. I'm your host, Jim Love. He is his CIO for IT World in Canada and TechNewsDay in the US.
It turns out that Elon Musk's lawsuit may have left out some details. A large part of Musk's lawsuit accuses OpenAI of changing from a nonprofit to a for-profit company and giving control of the AI to Microsoft.
However, documents published by OpenAI say the board did not want to become a for-profit company because a nonprofit would not be able to raise the funds needed to pursue the automated general intelligence mission. They also claim they knew all along that Musk would not object.
According to documents released by OpenAI, Musk not only did not oppose the deal, but also actively wanted a merger with Tesla, which he also owns.
The board was concerned about handing over control to a specific person, especially if that person is Elon Musk.
Regarding Musk's claims that the board waived its rights to artificial intelligence, making Microsoft the “de facto owner” of AGI, OpenAI already said in a staff memo yesterday that the deal with Microsoft does not give the company any ownership rights. He pointed out that it does not confer any ownership rights. Right to AGI. Additionally, the memo indicated that it is in the sole discretion of the board to state what is AGI and what is not.
“Standard” generative AI products created by OpenAI, such as GPT 4, are accessible. But everyone, including Musk, knew that without the $10 billion coming from Microsoft, OpenAI wouldn't have the computing power it needed. We will promote research and development of AI.
Or if you asked him to summarize Musk's position as if he were William Shakespeare, ChatGPT would say, “I think Elon protests too much.”
Source: AIGrid (YouTube) and the Verge
Recent takeaways from the Europol report were widely cited on the internet, including the statement that 90% of online content will be generated by AI by 2026.
The only problem is, according to an article published by a company called Oodalup, that quote is nowhere to be found in the report.
This article also includes some very interesting statistics from a report by UK-based company Public First.
62% of respondents supported the creation of a new government regulatory body, similar to the Healthcare Regulatory Authority, to regulate the use of new AI models.
Overall, 32% believe advanced AI will make them safer, while 18% think it will make them less safe. When asked about specific risks from advanced AI, increased unemployment is seen as the most important (49%)
Although this is a UK report, it contains some interesting research. A link will be posted in the program notes.
But as someone who puts together a daily news program, there was a real lesson here. Just because a lot of people say statistics exist doesn't mean they actually exist.
Source: Oodaloop
The personal and financial information of about 30,000 Fidelity Investments Life Insurance customers could be stolen in what appears to be a ransomware attack following a breach into Infosys' IT systems last fall. Compromised data could include bank account and routing numbers, credit card numbers and security codes, names, social security numbers, locations of residence, and dates of birth. This breach could potentially enable identity theft fraud and fraudulent financial transactions.
Indian technology services giant Infosys has experienced a cybersecurity incident affecting its US subsidiary Infosys McAmish Systems (IMS), leading to the outage of some applications and IT systems. His LockBit, a notorious ransomware group, claimed responsibility for the intrusion. The incident marks a new major data breach involving Infosys, following a similar disclosure last month related to the Bank of America data breach.
The cyber attack occurred between October 20 and November 2 and disrupted services provided by Infosys to both Fidelity and Bank of America. Fidelity has been working with her IMS to investigate and contain the event, implement corrective actions, and safely restore service.
Source: The Register
Hospital workers are exposed to up to 1,000 alarms per shift, creating a phenomenon known as “alarm fatigue.” This sensory overload is associated with hundreds of deaths annually. New research suggests that replacing standard monotonous hospital alarms with more musical alarms could significantly reduce alarm fatigue while making key equipment less cumbersome. Masu.
The study included experimenting with different musical sounds for hospital alarms, but found that only 15% of all alarms in critical care ward environments were clinically relevant. Ta. A high number of false alarms can contribute to alarm fatigue, or desensitization caused by sensory overload, which can cause alarms to be missed. Between 2005 and 2010, the US FDA reported 566 alert-related deaths.
Researchers Joseph Schlesinger, an anesthesiologist at Vanderbilt University Medical Center, and Michael Schutz, a music cognition researcher at McMaster University, explore how timbre can help busy healthcare workers with softer sounds. I have been researching whether it is possible to draw They found that “percussive” tonal sounds containing short bursts of high-frequency energy were noticeable even at low volumes. In contrast, loud “flat” tones without high frequency content tend to be lost.
In their study, 42 participants were given six alarms. One half was designed according to the standard alarm, and the other half had a new tone based on the sound of a xylophone. Researchers assessed participants' perceived discomfort with various alarms and their ability to recognize them. They found that in 88% of instances, complex percussion sounds were considered less unpleasant compared to standard sounds used in medical devices. The acoustically rich tone alarm melody was no more difficult to identify than a standard hospital beep.
This study shows that musical tones can significantly reduce discomfort without compromising alarm learning ability. This represents a step toward improving alarm design while addressing the problem of excessive alarm sounds among medical devices. Future research will investigate how different tones affect other important perceptual issues, such as alarm detectability.
Sources include: NewAtlas.com
The study was published in the journal Perioperative Care and Operating Room Management.
That's today's show.
Please like your comment.
Send a note to jlove@itwc.ca or drop a comment below the show notes at itworldcanada.com/podcasts. Look for hashtag trends.
Thank you for your attention. Have a thrilling Thursday!
