Google’s boss says AI and machine learning will help
谷歌老板说人工智能和机器学习将带来助益

绿色技术的力量  The power of green technology-书迷号 shumihao.com

AFTER DECADES of incremental steps forward, 2021 will be the most significant year yet for combating climate change. Two recent developments have made this possible.
经过数十年的小步进展,2021年将是应对气候变化最重要的一年。近期的两项进展使这成为可能。

First, as science tells us that we have a decade to reduce emissions dramatically or face the worst impacts of climate change, many of those impacts have already arrived at our door. From the historic and deadly wildfires in Australia and California, to severe flooding around the world, there is no denying that climate change is already disrupting our daily lives. At the same time, support for climate action has never been stronger—from Generation Z’s solutions-oriented mindset, to political support that increasingly crosses party lines, to Europe’s large-scale ambition to become the first carbon-neutral continent, society is ever more unified against the threat of climate change.
首先,正如科学告诉我们的那样,我们有十年的时间来大幅减少排放,否则会面临气候变化最严重的影响——其中许多影响已经到了我们的门前。从澳大利亚和加州历史性的致命野火,到世界各地的严重洪灾,气候变化已经在无可否认地破坏我们的日常生活。同时,对气候行动的支持从未如此强大——从Z世代以解决方案为导向的思维方式,到逐渐跨越党派的政治支持,再到欧洲要成为第一个碳中和大陆的宏大野心,全社会日益团结应对气候变化的威胁。

Second, we are seeing promising technologies and policies that will bring carbon-free energy within reach. Not long ago, it was hard to imagine a 24/7 carbon-free electricity supply. At its most basic level, the wind does not always blow and the sun does not shine at night. But new technologies—including better energy storage and the reduction of costs associated with wind and solar power by 70% and 89% respectively over the past ten years—are bringing 24/7 carbon-free energy closer to reality.
其次,我们看到了有前途的技术和政策,它们将使无碳能源触手可及。全天候无碳电力供应在不久前还难以想象。哪怕说最基本的,风并不是一直在吹,晚上也没有阳光。但过去十年来,新技术——包括更好的能量存储,以及与风能和太阳能相关的成本分别降低70%和89%——使全天候的无碳能源更加接近现实。

Another of those technologies is artificial intelligence (AI). At Google, we are working on ways to apply AI to optimise electricity consumption within our data centres. In collaboration with our sister venture, DeepMind, we have developed solutions that have reduced the amount of energy used to cool our data centres by 30%. This approach could be used by commercial buildings, including airports and shopping malls, to do the same. AI can also be used to make wind power more predictable, which will increase the value, utilisation and adoption of renewable energy.
这些技术中,另一个是人工智能(AI)。在谷歌,我们正在研究应用AI来优化数据中心内部电力消耗的方法。通过与我们的姊妹企业DeepMind合作,我们开发出了解决方案,将用于冷却数据中心的能源减少了30%。包括机场和购物中心在内的商业建筑都可以使用这种方法。AI还可以使风能更可预测,这将增加可再生能源的价值、利用率和接纳度。

Meanwhile, sensors on satellites can locate large-scale emitters of carbon dioxide at a very fine-grained level. This could dramatically improve the effectiveness of the Paris climate agreement. Technology is also helping cities reduce their carbon emissions. According to the Global Covenant of Mayors, an international alliance of over 10,000 cities and local governments committed to fighting climate change, less than 20% of cities outside western Europe have the time, resources and data to meet their climate commitments. With platforms like our own Environmental Insights Explorer, cities can use anonymised, aggregated mapping data to estimate the carbon footprint of their buildings and transport, and realise their solar-energy potential—a critical step, as cities continue to contribute over 70% of the world’s greenhouse-gas emissions.
同时,卫星上的传感器可以非常精细地定位大规模的二氧化碳排放源。这可以大大提高《巴黎气候协定》的有效性。技术也在帮助城市减少碳排放。根据《全球市长盟约》(由致力于抗击气候变化的一万多个城市和地方政府组成的国际联盟),西欧以外只有不到20%的城市有时间、资源和数据来履行其气候承诺。借助我们自己的环境洞察探索(Environmental Insights Explorer)之类的平台,城市可以使用匿名的汇总地图数据来估算其建筑物和交通的碳足迹,并实现其太阳能潜力——这是至关重要的一步,因为全世界70%以上的温室气体排放仍然源自城市。

Technology is also helping communities adapt to the effects of climate change that are already apparent. As one example, we are able to use satellite data to map wildfires in real time and better predict how they might spread. In India, flood forecasting models use AI to predict when floods will hit and how deep the waters will get, helping save lives. Machine learning is also being applied to “nowcast” rainfall sooner and with more accuracy than conventional forecasting methods, helping people make safer, more informed decisions.
技术还帮助社区适应已经显现的气候变化影响。举个例子,我们能够使用卫星数据实时绘制野火地图,并更好地预测野火的蔓延方式。在印度,洪水预报模型使用AI来预测洪水何时袭击、会达到多深,从而挽救生命。机器学习也可用于比传统的预测方法更快、更准确地“现报”降雨,从而帮助人们做出更安全、明智的决策。

Driven by these promising trends and tools, companies have made bigger sustainability commitments in shorter time frames. At Google, we have eliminated our carbon legacy using high-quality offsets, and set a goal to operate on 24/7 carbon-free energy in all our data centres and campuses worldwide by 2030. Our aim is to demonstrate that a 100% carbon-free electrical grid is not just possible but also economically viable. We hope that companies of all sizes will join us in this effort.
在这些有潜力的趋势和工具的推动下,企业纷纷做出了在更短时间内达成更大的可持续发展目标的承诺。在谷歌,我们已经使用高质量的补偿方案消除了碳遗留问题,并设定了到2030年在全球所有谷歌数据中心和园区全天候使用无碳能源的目标。我们的目标是证明100%的无碳电网不仅可能,而且在经济上可行。我们希望各种规模的公司都能加入我们的行列。

In addition to concrete and ambitious company commitments, the world also needs enabling policies and global frameworks to ensure we are working towards the same goals. We know it’s possible: we have seen this kind of collaboration during the pandemic, as the private sector worked with governments to deliver personal protective equipment, medical devices and contact-tracing apps needed to fight the virus. Stronger public-private partnerships will be just as critical in fighting climate change.
除了企业做出具体而宏伟的承诺外,世界还需要有利的政策和全球框架来确保我们正朝着相同的目标努力。我们知道这是可能的,在疫情期间已经出现了这种合作:私营部门与政府合作提供了抵抗病毒所需的个人防护设备、医疗设备和接触者追踪应用。加强公私合作对于应对气候变化同样至关重要。

Throughout history, every generation has confronted big challenges. Climate change will be our generation’s most profound challenge—and in 2021, the world will take its biggest steps yet to meet it.■
纵观历史,每一代人都面临巨大的挑战。气候变化将是我们这一代人最严峻的挑战——2021年,世界将迈出最大的步伐来应对它。