Dr John Coyne, Head of Strategic Policing and Law Enforcement, and Head of the North and Australia’s Security, says that it is important to have a solid understanding of the technology and tools that we use, along with the challenges they offer.
What is ASPI and how does it function?
The Australian Strategic Policy Institute (ASPI) is an independent think tank that is located in Canberra, Australia. We are an independent public policy think tank that works in the national security and defence space. ASPI was established 18 years ago by the Australian prime minister with bipartisan support from the opposition party. At the time the Australian government, and opposition, felt that they were not getting contested policy advice from the bureaucracy and the public servants. Hence, it decided to independently fund ASPI to provide the contested advice it needed.
We work across a range of national security issues. I cover strategic policing and law enforcement, borders and border management. Geospatial technology plays a significant role in national security, especially in maritime operations. It provides the means to understand what is happening in a nation’s maritime zone.
Developing a maritime domain awareness strategy involves figuring out what technologies and capabilities we can use to watch all of our sea and maritime operations for search and rescue, fisheries and checking smuggling or other illegal activities. I don’t use these technologies but I do provide advice to the government for decision and policymaking. My research program seeks to bring strategic partners together to ensure decisions on maritime domain awareness achieve the right balance of technology and capability. So, this puts us in an interesting position, where I spend a lot of time talking to ministers, working bureaucrats and military officials discussing surveillance in the maritime domain.
Has there been a change in the government’s response to geospatial and AI, and what are the challenges associated with these technologies?
Yes, there has been a change. I have been working with the Australian government for the past 25 years. In the past, governments had the technological edge, so it would create new capability, would use and classify. But in the last decade, that has changed significantly. Today, there are a lot more opportunities to engage with the private sector for surveillance technology, rather than develop it yourself. There is an opportunity to leverage public-private-partnership.
As the cost of these new capabilities is much lower, there are many new opportunities to engage with emerging technologies. Though there are a couple of challenges, one of them is that for AI to work, it needs a large data set, so the self-learning algorithm.
Also Read: Using latest tech for border management
In Australia and all of its waters, say we can find 115 illegal fishing boats every year, tis sample will guide self-learning algorithms: but is it truly representative. One of the challenge is how big the sample is. The second challenge is that Artificial Intelligence is only as good as the data that is going into the system. So, in terms of using it as a predictive tool, in some ways, AI can be self-adjusted and be as discriminating as a human being. A lot of people don’t understand that.
For instance, if the Malaysian police are probing a fraud angle for backpackers, and only search backpacker tourists that are dark-skinned and search them for drugs. If they find it, that goes into the self-learning algorithm that could then draw the conclusion that dark-skinned tourists are more likely to have drugs on them, which is incorrect. The third part of AI is that self-learning algorithms are have grown to such a level of complexity that actually understanding how they get to an answer is becoming more difficult.
We are reaching a point where AI that is self-learning, will keep on gathering new information. The algorithms behind law enforcement decision making will quickly become so complicated that police officers cannot actually describe how they got to a specific conclusion.
These are the challenging pieces with AI, around the region and around the world. On one side they have all the tech developers telling us that these products are amazing. It doesn’t matter if you are talking about geospatial, satellites, combining satellite data for surveillance, or AI.
The technology is changing at a fast pace, so its timely adoption is a challenge. It is an enabling technology, and in the end, you have to have manual intervention. Even as a tool, if you can’t understand how it works, it’s not going to help. As humanity becomes more complex, the data sets will only become larger.
In terms of geospatial analysis, I will say that it is really growing well. I think the technology in AI that sits behind this is going to be increasing. For instance, in London, there are tens of thousands of video or security cameras collecting information that goes wasted. More than 85% of all drone footage from the war in Afghanistan was never viewed by humans because we don’t have the system to combine that, take all of that data and do things like spatial recognition and pattern analysis. Looking at all of this is a hard thing. In terms of development, you have some really good startup companies doing things. The hard thing for them is they essentially got to sell to a client in order to scale. We too want the startup technology, but the government needs to be behind them to fund and that’s the challenge.
How do you use latest technology in your border security program?
I have a couple of ongoing major bodies of work at the moment. I am working the United Nations Office on drugs and crime, supporting the development of a roadmap for ASEAN security and economic integration. This body of work is focussed on achieving balance and harmony between streamlining minimum border security standards for border crossing across the whole of South East Asia.
We have got new connectivity from the Belt and Road Initiative in China, so you get new ports new roads, new train systems. You get people moving faster, more of them in groups, etc. The challenge in the region is how do we build up better security controls. Big Data and geospatial analysis are going to play a crucial role. And at the moment, on border roadmap, the key focus is on economic growth. Over the last twelve months, ASEAN’s member states have been much more proactive about dealing with the crises to avoid the number of irregular migrants in the region from growing.
Also Read: Adding innovation to law enforcement
https://ift.tt/2OjNWKP
0 Response to "The right mix of technology and capability for foolproof surveillance - Geospatial World"
Post a Comment