For the last nine years, Hugh Thompson has delivered one of the most anticipated and well-attended keynotes at the annual RSA Conference. This year's instalment of 'Hugh Thompson and Friends' started talking about monkeys and finished with conversations with a leading psychologist and a specialist in security metrics.
Thompson began with a story of a holiday in Bali. While living in Asia for half of last year, Thompson and his family visited the famed monkey forest and Thompson realised that the monkeys had developed a keen sense of which tourists would be prepared to give them more bananas to buy back items that the monkeys had stolen.
In a sense, this is what motivated and targeted attackers do – they look for the potential victims from whom they are more likely to gain a strong result.
Without ramming the point down the audience's throat, it was clear that Thompson was making a point. Attackers are motivated, smart and are looking for the low hanging fruit.
Thompson's first guest was Dr Angel Duckworth, a psychologist from the University of Pennsylvania who has developed a 12-question test that measures grit – passion and perseverance to achieve long-term goals. The aim was to find which students coming into West Point were more likely to graduate.
Although the questionnaire is relatively simple, it provides a way of predicting the level of commitment a candidate would put to an activity. But Duckworth found that the one quality that was the greatest predictor of ongoing success. This was follow-through.
Interestingly, Duckworth saw that grit in any activity was the strong predictor – the specific activity was not relevant. And she did note that it was people to develop grit by modelling the behaviour of grittier people.
"World class experts do deliberate practices. They are comfortable being uncomfortable," said Duckworth.
Dan Geer, Principal of Geer Risk Services, was Thompson's second guest. He has a focus on metrics around security. Thompson asked "Where do you think we are today with predictive analytics?"
"We're very close, in being able to predict better than you would have thought," said Geer. "The kind of data we are able to collect has surpassed our ability to look at it. That will not last long. There are lots of start-ups looking at exactly this. Nearly every leading edge security product has a behavioural component".
Thompson posited to Geer that one of the challenges facing the world is that data that was previously considered unimportant and was readily given away, data as diverse as Social Security Numbers and the names of pets, is now important. How do we create security models for data we don’t know is important?
"Re-purposing data is not something you can protect against.," replied Geer. "We can now do data fusion from multiple sources [using big data approaches]".
"I'm an convinced there is nearly no kind de-identification process that cannot be reversed if you have enough data," he added. "The right to be forgotten will never be achievable"
More RSA Content you maybe interested in: