I fear we may have created a monster - part 2

In the first of this two-part article, I discussed how the Internet, a tool that we conceived to bring so much good to the world, is increasingly being used as a weapon for bad.  In this concluding half I will further explore this, the significant role that Social Media plays, and how Artificial Intelligence will turbocharge things even further. 

The Internet has opened an entirely new theatre of operations for crime, giving global reach to petty criminals all the way through to large scale organised syndicates.  It is quite easy to disguise oneself, impersonate someone else, quickly retreat into the shadows, and take advantage of the complication of policing across international borders.

How misinformation spreads on social media—And what to do about it, Brookings Institute

Let’s now have a closer look at misinformation and disinformation.

Misinformation, i.e. false or inaccurate information spread without malicious intent, is nothing new, having been with us for thousands of years.  E.g. during the Great Plague there was plenty of misinformation saying it was caused by bad air, and hence could be avoided by burning herbs or incense, or by carrying posies of flowers.

Internet, and social media in particular, makes it easy for misinformation to spread with relatively little effort and cost to distribute to all corners of the globe.  Previously, information, misinformation and disinformation were relatively expensive to print and distribute in written form, and therefore had limited opportunity to spread beyond family or friend groups, towns, cities or countries.

Social media inflames the UK race riots, BBC

Disinformation is also nothing new, i.e. deceptive content deliberately created to mislead.  For example, Operation Mincemeat in World War two, where Allied forces successfully deceived the enemy into thinking they were going to invade Greece rather than Sicily.  Now in the Internet age, disinformation is spread at much greater scale by individuals, groups, and nations, all with their own agenda.  E.g. this Poynter Institute article describes Russia’s attempts to sow seeds of division in American society in an attempt to interfere in the 2016 USA election.  What makes this particularly scary is the audacity of the ambition and that the Internet, and social media in particular, is enabling manipulation of opinion through disinformation at a scale never seen before in human history.

Social Media weaponised

Social media is now one of the biggest weapons in crime, misinformation and disinformation.  Criminals can research organisations and individual employees to help them figure out who they could approach and how they could deceive.  For example, at a small company I worked at a few years ago, I received an email purportedly from the CEO.  The email said, “Hi Robert”, and then went on to say that he wanted to give some customers a gift card as a festive season gift.  As he was in meetings all day, could I please go out and purchase some gift cards, and send the activation codes from those cards to him.  This is a crude example of “spear phishing”, a form of phishing where criminals craft messages designed for specific individuals rather than generic messages.

Whilst it was easy to guess this was scam, every year there are many victims of more savvy spear phishing attempts where a criminal has spent more time researching their targets on LinkedIn, Facebook, or what-ever, and more time crafting a message that sounds authentic and that the target victim is likely to fall for.  According to a 2023 Barracuda Networks report, 50% of surveyed organisations were victims of spear phishing.

Generative AI turbocharges all this

Generative AI is the underpinning technology for the likes of ChatGPT, Microsoft Copilot, Google Gemini, and the imminent Apple Intelligence.  Amongst other things, it can write articles, create social media posts, and simulate conversations. 

Before generative AI, the number of fake personas and social media accounts was constrained by the amount of time it took to create and “operate” those accounts one by one.   By harnessing generative AI, one person or a small group of people can now create and operate 100’s or 1000’s of fake personas and social media accounts with a common purpose. 

It is increasingly difficult to tell the difference between a real human and a bot.  Using generative AI, these bots are capable of writing fairly convincing posts, and even responding to other people's reactions and responses to that bot's posts.   You could probably figure out what to say and ask to reveal that you are chatting with a bot and not a human … if it occurred to you to do that.  Inevitably, a significant proportion of people will be misled.  In many countries, it takes a few percent of the population to be influenced to cause a change of government.

Political and social influence has been with us for thousands of years.  The difference this time is that one person, or a small group of people, can create and orchestrate 1000's of pretend humans to spread their agenda.  This can be done at scale at low cost, without needing to be published or picked up and covered by traditional mass media.

What to know about US election hacking, Iran and other countries, BBC

After the attempt to assassinate Trump, president Biden called to lower the temperature in American politics.  How much of that temperature has been stoked by bot farms operated by individuals and even USA's adversaries?

The genie is out of the bottle, and it may prove to be a monster.

#cybersecurity #cyber #socialmedia #AI #smallbusiness