We develop a structural econometric model to capture the decision dynamics of human evaluators on an online micro-lending platform, and estimate the model parameters using a real-world dataset. We find two types of biases in gender, preference-based bias and belief-based bias, are present in human evaluators' decisions. Both types of biases are in favor of female applicants. Through counterfactual simulations, we quantify the effect of gender bias on loan granting outcomes and the welfare of the company and the borrowers. Our results imply that both the existence of the preference-based bias and that of the belief-based bias reduce the company's profits. When the preference-based bias is removed, the company earns more profits. When the belief-based bias is removed, the company's profits also increase. Both increases result from raising the approval probability for borrowers, especially male borrowers, who eventually pay back loans. For borrowers, the elimination of either bias decreases the gender gap of the true positive rates in the credit risk evaluation. We also train machine learning algorithms on both the real-world data and the data from the counterfactual simulations. We compare the decisions made by those algorithms to see how evaluators' biases are inherited by the algorithms and reflected in machine-based decisions. We find that machine learning algorithms can mitigate both the preference-based bias and the belief-based bias.

This paper explores the environmental impact of the super-linear growth trends for AI from a holistic perspective, spanning Data, Algorithms, and System Hardware. We characterize the carbon footprint of AI computing by examining the model development cycle across industry-scale machine learning use cases and, at the same time, considering the life cycle of system hardware. Taking a step further, we capture the operational and manufacturing carbon footprint of AI computing and present an end-to-end analysis for what and how hardware-software design and at-scale optimization can help reduce the overall carbon footprint of AI. Based on the industry experience and lessons learned, we share the key challenges and chart out important development directions across the many dimensions of AI. We hope the key messages and insights presented in this paper can inspire the community to advance the field of AI in an environmentally-responsible manner.

Some of the excesses of 2021 have shown us how digital technologies can undermine what philosophers call future "human flourishing." A lot has been written on this topic in the first few days of the new year, but take two examples -- MIT Technology Review's list of the worst excesses of technology and Fast Company's 5 best and worst tech moments of 2021 -- and it's evident how little power people affected by technologies have when things go wrong under current systems. What's also clear as we enter 2022 is that global tolerance for technology's unchecked disruption of societal institutions, conventions, and values is waning. This is the year governments will pass legislation to control the effects of digital technologies on societies, across many jurisdictions and in relation to numerous existing and emergent technologies. The EU AI and Digital Services Acts, the UK Online Safety Bill, and the US SAFE TECH Act are just a few of the efforts underway. Legislation is a marker of societal concern, but it's also clear that non-specialist, "ordinary" people have an increasingly sophisticated understanding of the relationship between technology and society.

It's 2022, and developments in the AI industry are off to a slow -- but nonetheless eventful -- start. While the spread of the Omicron variant put a damper on in-person conferences, enterprises aren't letting the pandemic disrupt the course of technological progress. John Deere previewed a tractor that uses AI to find a way to a field on its own and plow the soil without instructions. As Wired's Will Knight point outs, it and -- self-driving tractors like it -- could help to address the growing labor shortage in agriculture; employment of agriculture workers is expected to increase just 1% from 2019 to 2029. But they also raise questions about vendor lock-in and the role of human farmers alongside robots.

Over the past six months, the Chinese government has rolled out a series of policy documents and public pronouncements that are finally putting meat on the bone of the country's governance regime for artificial intelligence (AI). Given China's track record of leveraging AI for mass surveillance, it's tempting to view these initiatives as little more than a fig leaf to cover widespread abuses of human rights. Anyone who wants to compete against, cooperate with, or simply understand China's AI ecosystem must examine these moves closely.

Tesla's "full self-driving" (FSD) feature has had something of a rocky history, to put things generously. It's been implicated in multiple crashes, seemingly persistent software bugs and a cavalcade of scrutiny from a panoply of regulatory bodies. Also, it's about to cost more money. The bump represents an additional $2,000 being tacked onto the not insubstantial price tag: a new grand total of $12,000, or most of the way to a Honda Civil. Nor is it the first time FSD has gotten more expensive.

Artificial intelligence should be carefully regulated as of now since the concept is still broad. What should be heavily regulated is its applications including autonomous driving, cybersecurity and the military. It's way too early to regulate a fundamental technology such as artificial intelligence. If you ask any expert as of today what should be regulated in AI, the answer would have to be, inevitably, "we don't know". While the rapid progress of the technology should be seen with a positive lens, it is important to exercise some caution and introduce laws that will help the progress of AI technology.

In 2019 Estonia set about designing a "robot …

Unlike many valuable resources, real-time data is both abundant and growing rapidly. But it also needs to be handled with great care. That was one of the key takeaways from an online workshop produced by Science Business' Data Rules group, which explored what the rapid growth in real-time data means for artificial intelligence (AI). Real-time data is increasingly feeding machine learning systems that then adjust the algorithms they use to make decisions, such as which news item to display on your screen or which product to recommend. "With AI, especially, you want to make sure that the data that you have is consistent, replicable and also valid," noted Chris Atherton, senior research engagement officer at GÉANT, who described how his organisation transmits data captured by the European Space Agency's satellites to researchers across the world.

As CES wraps up, we're still pulling together our favorite picks of the show. That includes finger-nibbling robots, smart beds and all kinds of TVs, laptops and gadgets. Yes, we've been able to see some of the products while not attending the show, but it has meant a lot of spec-sheet perusing and a fair dose of skepticism without getting a lot of the announcements in the flesh. For things like TVs, that's usually months later, but for tablets, phones and wearables, you can expect Engadget to be reviewing and stress-testing many of them sooner rather than later. Have a great weekend and see you back here Monday.