As the data center backlash grows, support is growing for server factories and the hundreds of jobs they're expected to bring. Last month, Pamela Griffin and two other residents of Taylor, Texas, took to the lectern at a city council meeting to object to a data center project. But later, they sat back as council members discussed a proposed tech factory. Griffin didn't speak up against that development. A similar contrast is repeating in communities across the US.

In this paper, we investigate the impact of numerical instability on the reliability of sampling, density evaluation, and evidence lower bound (ELBO) estimation in variational flows. We first empirically demonstrate that common flows can exhibit a catastrophic accumulation of error: the numerical flow map deviates significantly from the exact map---which affects sampling---and the numerical inverse flow map does not accurately recover the initial input---which affects density and ELBO computations. Surprisingly though, we find that results produced by flows are often accurate enough for applications despite the presence of serious numerical instability. In this work, we treat variational flows as chaotic dynamical systems, and leverage shadowing theory to elucidate this behavior via theoretical guarantees on the error of sampling, density evaluation, and ELBO estimation. Finally, we develop and empirically test a diagnostic procedure that can be used to validate results produced by numerically unstable flows in practice.

Feedback is New Scientist's popular sideways look at the latest science and technology news. You can submit items you believe may amuse readers to Feedback by emailing feedback@newscientist.com Feedback has been dimly aware for a while that there is a slew of AI-generated music swamping platforms like Spotify. Our awareness was limited, we confess, because we are so old that we still prefer to listen to CDs. Still, we weren't too surprised when New Scientist's Timothy Revell told us about an indie rock band called The Velvet Sundown that appears to be entirely AI-generated, from their songs, which sound like the beige love-children of Coldplay and the Eagles, to their uncanny-valley Instagram photos, which look like rejected concept art from Daisy Jones & the Six.

In this paper, we investigate the impact of numerical instability on the reliability of sampling, density evaluation, and evidence lower bound (ELBO) estimation in variational flows. We first empirically demonstrate that common flows can exhibit a catastrophic accumulation of error: the numerical flow map deviates significantly from the exact map---which affects sampling---and the numerical inverse flow map does not accurately recover the initial input---which affects density and ELBO computations. Surprisingly though, we find that results produced by flows are often accurate enough for applications despite the presence of serious numerical instability. In this work, we treat variational flows as chaotic dynamical systems, and leverage shadowing theory to elucidate this behavior via theoretical guarantees on the error of sampling, density evaluation, and ELBO estimation. Finally, we develop and empirically test a diagnostic procedure that can be used to validate results produced by numerically unstable flows in practice.

It's hard to overstate the impact of artificial intelligence (AI) on the world, and few are better equipped to discuss its potential than Microsoft co-founder Bill Gates.

Artificial intelligence (AI) and automation are rapidly changing the way we work and live, and it's no surprise that many are wondering how these technologies will affect employment opportunities. As machines take on more tasks traditionally performed by humans, it's natural to ask what the future of work will look like. One potential impact of AI and automation on the workforce is job displacement. Some experts estimate that up to 50% of all jobs could be automated in the next decade or two. While this may seem alarming, it's important to remember that this process has been happening for centuries and has often led to the creation of new and more advanced jobs.

The world is about to get very uncomfortable for a lot of companies. The next few decades will bring us not just one paradigm shift as impactful as the rise of the internet -- but 5 to 10 of them, coming online far faster than that "polite" revolution. Companies aren't built for the scope and speed of the coming disruption. Already, the old corporate guard has been crumbling under its own bureaucracy. Yet even new, rising-star companies may grow fast, then fail to adapt to the next disruption, because they're not prepared to move at the new speed of change.

Have you used the customer service app with your bank the past year, or received an unexpected email with an offer you were actually interested in? Maybe it was a well-timed mortgage re-fi or even some savings at a favorite store. There's an enduring and unfortunate misperception that AI serves only to replace human workers and irritate human clients. But what if we don't have to be locked in a zero-sum game with the growing legion of digital intelligence? An increasing number of banks and insurers are finding that it's almost impossible to meet the rising customer expectations and needs that digital services and apps have unleashed.

Digital disruption has broken out of Silicon Valley. Any company, no matter how nuts-and-bolts, can be disrupted by a digital competitor; equally, any company could be that digital disruptor. The discussion was kick-started by two leading industry thinkers: Andrew Moore, chief transformation officer of chipmaking giant Intel, and Nigel Moulton, chief technology officer at Dell EMC, part of a corporation that services 99% of the Fortune 500 companies. Their remarks sparked lively discussion. Both Intel's Mr Moore and Dell EMC's Mr Moulton spend a lot of time talking to leading companies about their digital transformation journey, and they kicked off with a tough message: it's hard work.

Digital disruption has broken out of Silicon Valley. Any company, no matter how nuts-and-bolts, can be disrupted by a digital competitor; equally, any company could be that digital disruptor. The discussion was kick-started by two leading industry thinkers: Andrew Moore, chief transformation officer of chipmaking giant Intel, and Nigel Moulton, chief technology officer at Dell EMC, part of a corporation that services 99% of the Fortune 500 companies. Their remarks sparked lively discussion. Both Intel's Mr Moore and Dell EMC's Mr Moulton spend a lot of time talking to leading companies about their digital transformation journey, and they kicked off with a tough message: it's hard work.