The UK’s unemployment rate rose to 4.3% in the three months leading up to September, marking an increase from 4% in the previous quarter, according to recent data from the Office for National Statistics (ONS). However, the ONS has cautioned against relying heavily on these figures due to data collection challenges that may impact accuracy.

Despite the slowing of wage growth, pay increases remain above inflation, which measures the rate of price rises. Data reveals that excluding bonuses, wages grew by an annual rate of 4.8% between July and September, the slowest rate in over two years. This deceleration could reflect the easing of the labour market, where job demand has waned. Vacancy numbers have been consistently declining for more than two years, indicating a broad softening in hiring trends.

Liz McKeown, ONS director of economic statistics, highlighted on the BBC’s Today programme that while the latest data hints at a “continued easing of the labour market,” the response rate for the ONS’s Labour Force Survey—a primary source of UK employment data—has been lower than usual, raising concerns about the data’s reliability. The accuracy of these figures is crucial as the Bank of England closely monitors employment data when making interest rate decisions. Last week, the Bank reduced rates for the second time this year, following a drop in inflation to 1.7%, below its 2% target.

McKeown acknowledged that issues with the current data are affecting the Bank’s decision-making process, and efforts are underway to refine the methodology.

Adding to businesses’ challenges, a rise in National Insurance contributions (NICs) and the upcoming minimum wage increase have sparked concerns from major employers. Supermarket chains, including Asda and Sainsbury’s, along with High Street retailer Marks and Spencer, have expressed concerns about rising operational costs due to these new measures, announced in Chancellor Rachel Reeves’ recent Budget.

For businesses like Wendy Jones-Blackett’s greeting card company, headquartered near Leeds, these changes bring a potential strain on finances. Jones-Blackett mentioned that rising costs for outsourced services like printing and storage might impact her company’s ability to provide competitive pay increases. “If you want to retain good staff, you need to increase their pay,” she explained. “We want to do that but will have to temper it with rising costs.”

On the hiring front, recent data from the Recruitment and Employment Confederation and consultancy KPMG show that job vacancies have dropped for a consecutive 12 months, underscoring a reduced demand for new workers. Alexandra Hall-Chen, a principal policy adviser at the Institute of Directors, warned that the tax increases and new employment regulations introduced in the Employment Rights Bill are putting significant strain on hiring plans. “The cumulative effect of these changes will ultimately stifle job creation,” she said, urging the government to address businesses’ concerns.

Rob Wood, chief UK economist at Pantheon Macroeconomics, noted that while unemployment may gradually increase, the Bank of England is expected to focus on broader trends rather than specific data fluctuations when assessing future interest rate changes.

Meanwhile, Work and Pensions Secretary Liz Kendall has emphasized the importance of improving living standards, noting that three million of the lowest-paid workers are set to benefit from a minimum wage increase starting in April.

OpenAI, the company behind the AI chatbot ChatGPT, has rejected more than 250,000 requests to generate images of key US election candidates using its platform, DALL-E. The rejections, which were disclosed in a company blog update on Friday, were part of measures aimed at ensuring the safety and integrity of the upcoming election period.

Requests for AI-generated images of high-profile figures like president-elect Donald Trump, his vice-presidential pick JD Vance, current president Joe Biden, vice-presidential candidate Kamala Harris, and Tim Walz were all blocked. According to OpenAI, these refusals were implemented as part of “safety measures” to prevent the platform from being used to create misleading or harmful content in the lead-up to Election Day.

“These guardrails are especially important in an elections context and are a key part of our broader efforts to prevent our tools being used for deceptive or harmful purposes,” the blog post explained. The company emphasized that it had not seen evidence of any widespread influence operations in US elections through its platforms.

In addition to the image rejections, OpenAI revealed that it had taken action earlier this year against a political influence campaign linked to Iran. In August, OpenAI blocked the Iranian campaign, known as Storm-2035, from generating political content on US elections, which attempted to impersonate both conservative and progressive news outlets. Accounts tied to this campaign were subsequently banned from using OpenAI’s services.

The company also highlighted in an October update that it had disrupted more than 20 separate influence operations and deceptive networks from around the world that had attempted to use OpenAI tools for misleading purposes. However, the company’s report noted that none of these election-related operations managed to generate significant “viral engagement.”

OpenAI’s proactive steps to protect the integrity of its platforms during the election period are part of its broader efforts to minimize the potential for AI-generated content to be used maliciously or deceptively, especially in politically sensitive contexts. Despite concerns about the potential misuse of AI technologies for creating misinformation, OpenAI asserts that its safety measures are working to limit such risks.