British military intel chief has ‘ambitious plans’ for automation

LONDON — British defense analysts have access to more data and information sources than ever before, but are desperate for mature automation tools to properly sift through them all, said the nation’s top intelligence officer.

The U.K.’s defense intelligence arm has “ambitious plans” to increase the use of automation systems across its enterprise, and relieve its human analysts of performing repetitive tasks, said British Army Lt. Gen. James Hockenhull, chief of defense intelligence.

“Open source intelligence and automation together will fundamentally change how intelligence operates,” he said Thursday at the biennial DSEI conference. “We’re going to need access to data of all sorts, commercial expertise to help us to build our own open-source tools, and the ability to buy open source-derived commercial assessments.”

For several years now, intel officials have looked for solutions that can turn vast troves of information into manageable portions for analysis. That’s still the case today, but the requirements for such solutions have changed, Hockenhull noted.

Open source information gathering has come a long way from the early days of social media scraping, and “treasure trays” abound from the dark web and financial data resources, “if you can manage the volumes of data,” he said.

That last point is key: “If you talk to our analysts, many will wince at the idea of even more information and intelligence,” he noted. What they ask for is “a way to manage the deluge of data.”

To tackle that problem, the MoD is investing in mature automation, machine learning, and data science solutions, and wants to work with industry to further refine and hone more sophisticated artificial intelligence systems, Hockenhull said.

“As real artificial intelligence matures, I know it will play a vital role in our future. But it will also bring with it a host of challenges around exploitability, fragility and ethics,” Hockenhull said. “Our focus, at least initially within defense intelligence, will be on exploiting the more mature technologies, while ensuring that we play our part in helping shape the future ones.”

Today, he’s looking for data sets with open interfaces, rather than pre-packaged platforms, tools and data.

“As we mature our information systems, there will be less willingness to buy bespoke architectures or standalone platforms,” he noted. He said he envisions an open source intelligence automation system as “a blend” of tooling built in-house, bespoke software for specific needs and industry support to match, along with commercial-off-the-shelf products.

Space capabilities are being more closely scrutinized, he added. “With new commercial space collection [techniques] and other novel approaches coming online daily, this is something we can no longer dismiss, perhaps, as we have done in the past.”

To accomplish these goals, the MoD and its industry partners need to find new ways of working together, he acknowledged. “This is so difficult to achieve with hand-to-mouth annual contracting.”

In an effort to engage more fruitfully with the private sector, the defense intelligence unit is hosting an industry day on Oct. 14, he noted.

Vivienne Machi is a reporter based in Stuttgart, Germany, contributing to Defense News' European coverage. She previously reported for National Defense Magazine, Defense Daily, Via Satellite, Foreign Policy and the Dayton Daily News. She was named the Defence Media Awards' best young defense journalist in 2020.

伦敦——英国最高情报官员表示，英国国防分析人员现在比以往任何时候都能获取更多的数据和信息来源，但他们迫切需要成熟的自动化工具来正确地筛选所有这些数据和信息。

英国国防情报部门负责人、陆军中将詹姆斯·霍肯霍尔表示，英国国防情报部门制定了“雄心勃勃的计划”，旨在提高整个机构自动化系统的使用率，并使分析人员摆脱重复性工作。

“开源情报和自动化相结合将从根本上改变情报运作方式，”他周四在两年一度的DSEI大会上表示。“我们需要获取各种类型的数据，需要商业专业知识来帮助我们构建自己的开源工具，还需要能够购买基于开源的商业评估报告。”

多年来，情报官员一直在寻找能够将海量信息转化为易于分析的片段的解决方案。霍肯霍尔指出，如今情况依然如此，但对这类解决方案的要求已经发生了变化。

他说，开源信息收集已经从早期的社交媒体抓取发展到了今天，暗网和金融数据资源中充满了“宝藏”，“前提是你能够管理海量数据”。

最后一点至关重要：“如果你和我们的分析师交谈，许多人一想到要获取更多信息和情报就会感到畏缩，”他指出。他们真正需要的是“一种管理海量数据的方法”。

霍肯霍尔表示，为了解决这个问题，英国国防部正在投资成熟的自动化、机器学习和数据科学解决方案，并希望与业界合作，进一步完善和改进更复杂的人工智能系统。

霍肯霍尔表示：“随着人工智能的日趋成熟，我知道它将在未来发挥至关重要的作用。但它也会带来一系列挑战，例如易被滥用、脆弱性和伦理问题。至少在国防情报领域，我们的重点最初将放在利用更成熟的技术上，同时确保我们也能为塑造未来的技术贡献力量。”

如今，他寻找的是具有开放接口的数据集，而不是预先打包好的平台、工具和数据。

他指出：“随着信息系统的日趋成熟，人们购买定制架构或独立平台的意愿将会降低。” 他设想的开源智能自动化系统将是“一种融合体”，它结合了内部开发的工具、针对特定需求定制的软件、相应的行业支持以及现成的商用产品。

他补充说，太空能力正受到更密切的审视。“随着新的商业太空数据采集技术和其他创新方法每天都在涌现，我们或许不能再像过去那样对此置之不理了。”

他承认，为了实现这些目标，国防部及其行业合作伙伴需要找到新的合作方式。“靠捉襟见肘的年度合同很难做到这一点。”

他指出，为了更有效地与私营部门进行合作，国防情报部门将于 10 月 14 日举办行业日活动。

薇薇安·马奇是驻德国斯图加特的一名记者，主要负责《防务新闻》的欧洲报道。她此前曾为《国家防务杂志》、《防务日报》、《卫星通讯》、《外交政策》和《代顿每日新闻》撰稿。2020年，她荣获防务媒体奖最佳青年防务记者称号。