The Guardian · 1 May, 05:00
A study found that large-scale sporting events, such as the Commonwealth Games, can cause significant air pollution, primarily due to catering activities. The research, conducted during the 2022 Commonwealth Games in Birmingham, revealed that particle pollution in fan areas around the stadium more than doubled spectators' daily exposure.
Researchers from the University of Birmingham measured air pollution during the 2022 Commonwealth Games, which took place in Birmingham, UK. They used an electric van equipped with instruments to measure pollution levels around the Alexander Stadium, where athletics events and ceremonies were held. The study found that particle pollution, primarily caused by catering activities such as fast-food cooking, peaked before stadium sessions and reached a maximum just before the closing ceremony. The pollution levels were 10 times greater than in other parts of the city.
Why it matters: The study's findings are significant because they highlight the importance of considering air pollution impacts when organizing large-scale sporting events. Poor air quality can affect not only spectators but also athletes' performance and staff. The research suggests that event organizers should take steps to mitigate pollution from cooking activities at concession stands.
6,000 athletes, 72 countries, 3 million people, 300,000 spectators, 10 times greater pollution levels, 34 venues with capacity over Alexander Stadium
Event organisers should consider the impact of cooking from concession stands, particularly for staff and volunteers who may be exposed for multiple days. - Prof William Bloss
The Guardian · 1 May, 05:00
Many UK job seekers have expressed frustration with AI interviews, citing a lack of human interaction and difficulties in communicating naturally. Research found that 47% of UK job seekers have had an AI interview, and 30% have walked away from a hiring process because it included an AI interview.
A survey of 2,950 active job seekers by Greenhouse found that 47% of UK job seekers have had an AI interview. Many respondents shared their negative experiences with AI interviews, describing them as 'awkward', 'humiliating', and 'completely horrible'. Some found it difficult to speak naturally and felt that the AI interviews lacked a human element. Others were unsure if their interviews were even reviewed by a human.
Why it matters: The prevalence of AI interviews and the negative experiences shared by job seekers highlight the need for companies to improve their approach to AI interviews. The lack of human interaction and difficulties in communicating naturally may lead to candidates walking away from the hiring process, potentially resulting in missed opportunities for both the candidate and the company.
47% of UK job seekers have had an AI interview, 30% of UK candidates have walked away from a hiring process because it included an AI interview, 2,950 active job seekers surveyed, 1,132 UK-based workers surveyed.
AI interviews are one-way. They minimise the investment for the hiring party and maximise the strain on the potential supplier. They're also completely horrible for the autistic brain, and I presume not much nicer for others. - David*, 47, marketing consultant
The Guardian · 1 May, 05:00
In the upcoming UK local, mayoral, and devolved elections on May 7, men are expected to dominate the ballot papers, with women making up only a third of candidates. This underrepresentation of women is significant, as it may lead to their views and perspectives being sidelined in local and national decision-making.
According to an analysis by 50:50 Parliament and Democracy Club, in the upcoming UK local, mayoral, and devolved elections on May 7, women make up only a third of candidates, while men make up two-thirds. The analysis of 26,813 candidates found that no party achieved gender parity, with Labour being the closest with 42% female candidates. The research also found that in local elections in England, 34% of candidates are female, while in mayoral elections, the figure is 18%.
Why it matters: The underrepresentation of women on ballot papers is significant because it may lead to their views and perspectives being sidelined in local and national decision-making. This can have a negative impact on issues that affect women, such as social care and healthcare. Campaigners argue that this imbalance can affect the quality of democracy and decision-making.
A third of candidates are women, two-thirds are men; 34% of local election candidates in England are female; 18% of mayoral election candidates are female; 42% of Labour candidates are women; 516 Davids, 453 Johns, 385 Pauls, and 137 Sarahs among the candidates; 51% of the population are women, but only 41% of MPs are women; 26,813 total candidates analyzed.
We are 51% of the population, and our needs and experiences are not going to be represented by an endless supply of Davids and Johns. - Penny East, chief executive of the Fawcett Society
Ars Technica · 30 Apr, 21:11
Elon Musk took the stand for the third day in a trial against OpenAI, where he alleged the company abandoned its mission. However, Musk made several stumbles on the stand, including concessions, contradictions, and displays of dishonesty or disingenuousness.
Elon Musk testified for over seven hours in a trial against OpenAI, which he sued for allegedly abandoning its mission and planning to go public. Musk claimed OpenAI executives 'stole a charity' and that he was conned into giving 'free funding' to start a nonprofit. However, during his testimony, Musk made several mistakes, including concessions over his own lawyer's objections, losing a fight to keep xAI's safety record off the table, and appearing dishonest or disingenuous at times.
Why it matters: Musk's stumbles on the stand may jeopardize his chances of winning the trial, which could have significant implications for OpenAI's future plans, including its potential IPO later this year. If Musk loses, Sam Altman may remain at the helm of OpenAI, and the company may not remain a nonprofit forever.
$38 million, $800 billion
Ars Technica · 30 Apr, 22:09
President Trump has withdrawn his nomination of Casey Means for surgeon general and nominated Nicole B. Saphier, a breast radiologist and Fox News contributor, instead. Saphier has questioned vaccines and founded an herbal supplement company.
President Trump announced on social media that he was withdrawing his nomination of Casey Means to be surgeon general. Means' nomination had stalled in the Senate due to concerns about her vaccine views and qualifications. Trump then announced a new nominee, Nicole B. Saphier, a breast radiologist at Memorial Sloan Kettering Cancer Center and a Fox News contributor. Saphier has questioned vaccines and founded an herbal supplement company.
Why it matters: The nomination of a new surgeon general is significant as it affects the country's top doctor and their role in advising on health matters. The previous nominee, Casey Means, faced concerns about her qualifications and vaccine views, which may have contributed to her nomination being stalled. The new nominee, Nicole B. Saphier, also has unconventional views on vaccines.
Ars Technica · 30 Apr, 20:20
A severe Linux vulnerability, known as CopyFail (CVE-2026-31431), has been publicly disclosed along with exploit code that allows for easy elevation of privileges to root access across virtually all Linux releases. This has set off alarm bells as defenders scramble to patch the vulnerability and prevent severe compromises.
The vulnerability, CopyFail (CVE-2026-31431), is a local privilege escalation flaw that allows unprivileged users to gain root access. It was privately disclosed to the Linux kernel security team five weeks prior to the public release of the exploit code by researchers from Theori. The Linux kernel team has patched the vulnerability in several versions of the Linux kernel, but many Linux distributions had not incorporated these fixes at the time the exploit was made public.
Why it matters: This vulnerability is particularly severe because it can be exploited with a single piece of exploit code that works across all vulnerable distributions without modification. This allows for easy hacking of multi-tenant systems, breaking out of containers, and creating malicious pull requests through CI/CD workflows.
CVE-2026-31431, Linux kernel versions 7.0, 6.19.12, 6.18.12, 6.12.85, 6.6.137, 6.1.170, 5.15.204, and 5.10.254
NPR News · 1 May, 05:41
Summary not available.
NPR News · 1 May, 05:19
Summary not available.
NPR News · 1 May, 05:04
Summary not available.