Schneier on Security

AUGUST 15, 2023

This is why we need regulation: Zoom updated its Terms of Service in March, spelling out that the company reserves the right to train AI on user data with no mention of a way to opt out. Zoom execs swear the company won’t actually train its AI on your video calls without permission, even though the Terms of Service still say it can.

Technology 238

Technology Artificial Intelligence Surveillance 238

Schneier on Security

DECEMBER 19, 2023

Dropbox isn’t sharing all of your documents with OpenAI. But here’s the problem: we don’t trust OpenAI. We don’t trust tech corporations. How many more turned off that AI toggle, ruling out ever evaluating if those features were useful for them or not? It seems not to be true. We have no reason to.

Government 254

Government Banking Risk Internet 254

Daniel Miessler

MAY 15, 2023

Introduction Purpose Components Attacks Discussion Summary Introduction This resource is a first thrust at a framework for thinking about how to attack AI systems. There has been, of course, much content on attacking pre-ChatGPT AI systems, namely how to attack machine learning implementations. So things are very early.

Technology 364

Technology Internet Internet Mobile 364

Schneier on Security

SEPTEMBER 8, 2023

This was just one milestone in the race across industry and academia to find the best ways to teach LLMs how to manipulate tools, which would supercharge the potential of AI more than any of the impressive advancements we’ve seen to date. But most AI tools can’t place the order, not even online.

Banking 238

Banking Risk Accountability Government 238

Schneier on Security

AUGUST 22, 2023

License plate scanners aren’t new. What’s new is that AI is being used on the data, identifying “suspicious” vehicle behavior: Typically, Automatic License Plate Recognition (ALPR) technology is used to search for plates linked to specific crimes. The AI was not just looking at license plates either.

Surveillance 198

Surveillance Mobile Technology Artificial Intelligence 198

Schneier on Security

JULY 25, 2023

The details are scant—the article is based on a “heavily redacted” contract—but the New York subway authority is using an “AI system” to detect people who don’t pay the subway fare. A police spokesperson declined to comment.

Artificial Intelligence 178

Artificial Intelligence 178

Krebs on Security

APRIL 9, 2024

Another interesting bug McCarthy pointed to is CVE-2024-29063 , which involves hard-coded credentials in Azure’s search backend infrastructure that could be gleaned by taking advantage of Azure AI search. “Microsoft has updated their backend and notified any customers who have been affected by the credential leakage.”

DNS 234

DNS Social Engineering Malware Media 234

Troy Hunt

FEBRUARY 4, 2024

link] — Troy Hunt (@troyhunt) January 22, 2024 What I didn't know at the time was the hilarity of how similar this service would be to those that had come before it. Either way, collecting and storing this data is now trivial so not a big surprise to see someone screw up their permissions and (re)leak it all.

Marketing 224

Marketing 224

Schneier on Security

MAY 25, 2023

We don’t know because OpenAI doesn’t talk about their processes, how they validate the prompts they use for training, how they vet their training data set, or how they fine-tune ChatGPT. Their secrecy means we don’t know if ChatGPT has been safely managed.

Artificial Intelligence 245

Artificial Intelligence Hacking 245

Schneier on Security

MAY 11, 2023

We will all soon get into the habit of using AI tools for help with everyday problems and tasks. Imagine you’re using an AI chatbot to plan a vacation. Later, when you’re using another AI chatbot to learn about a complex economic issue, is the chatbot reflecting your politics or the politics of the company that trained it?

Artificial Intelligence 286

Artificial Intelligence Computers and Electronics Risk Technology 286

Schneier on Security

DECEMBER 5, 2023

AI is about to change that. Summarization is something a modern generative AI system does well. The technologies aren’t perfect; some of them are pretty primitive. And, unlike humans, AI tools can be replicated by the millions and are improving at astonishing rates. Spying is limited by the need for human labor.

Surveillance 331

Surveillance Internet Internet Government 331

Troy Hunt

APRIL 22, 2024

Don’t get pwned; get real-time alerts & prevent breaches #SecureYourSite Le Slip Français was breached by "shopifyGUY" (I wonder where all these Shopify API keys are coming from?!) The end result is greater awareness and a longer backlog of breaches to process than I've ever had before!

Data breaches 212

Data breaches Accountability Ransomware Hacking 212

Schneier on Security

JANUARY 26, 2024

One direction that these chatbots may lead us in is toward a society where we ascribe humanity to AI systems, whether abstract chatbots or more physical robots. This isn’t new: People projected intelligence and empathy onto the very primitive 1960s chatbot, Eliza.)

Advertising 261

Advertising Media Marketing Technology 261

Schneier on Security

JUNE 1, 2023

Earlier this week, I signed on to a short group statement , coordinated by the Center for AI Safety: Mitigating the risk of extinction from AI should be a global priority alongside other societal-scale risks such as pandemics and nuclear war. I actually don’t think that AI poses a risk to human extinction.

Risk 280

Risk Artificial Intelligence Technology Internet 280

Troy Hunt

MARCH 24, 2024

As an example, I asked ChatGTP about the validity of the pipe symbol during the live stream and according to the AI, it's permissible "when properly quoted": "john|doe"@example.com Give that a go and see how far you get in an input of type "email" Mind you, that example allows a pipe when not quoted.

217

217

Schneier on Security

DECEMBER 5, 2023

AI is about to change that. Summarization is something a modern generative AI system does well. The technologies aren’t perfect; some of them are pretty primitive. And, unlike humans, AI tools can be replicated by the millions and are improving at astonishing rates. Spying is limited by the need for human labor.

Surveillance 287

Surveillance Government Advertising Marketing 287

The Last Watchdog

AUGUST 13, 2023

LAS VEGAS – Just when we appeared to be on the verge of materially shrinking the attack surface, along comes an unpredictable, potentially explosive wild card: generative AI. Unsurprisingly, generative AI was in the spotlight at Black Hat USA 2023 , which returned to its full pre-Covid grandeur here last week.

Artificial Intelligence 246

Artificial Intelligence Engineering Internet Internet 246

Joseph Steinberg

NOVEMBER 1, 2023

While many people seem to be discussing the dangers of Artificial Intelligence (AI) – many of these discussions seem to focus on, what I believe, are the wrong issues.

Artificial Intelligence 169

Artificial Intelligence Social Engineering Engineering Phishing 169

Security Boulevard

DECEMBER 13, 2023

The use of generative AI systems has been spreading like wildfire, and if systems are not developed securely, the blaze could end up burning your organization. To help organizations tackle the problem, t he United Kingdom's National Cyber Security Centre (NCSC) and the U.S.

Software 121

Software Cybersecurity 121

Graham Cluley

MARCH 27, 2024

Security researchers find a way to unlock millions of hotel rooms, the UK introduces cyberflashing laws, and Google's AI search pushes malware and scams.

Scams 100

Scams Hacking Malware Cybersecurity 100

Jane Frankland

MARCH 27, 2024

And I can’t help but be concerned that technology, especially advancements in artificial intelligence, is not just reshaping our world; it’s actually reshaping our identity and the fabric of trust that binds us. If we are to thrive in this AI-augmented reality, we must not lose sight of what it is to be human.

Technology 130

Technology Artificial Intelligence Authentication Scams 130

Schneier on Security

AUGUST 23, 2023

We don’t have any legacy systems from the US or any other country. We don’t have any special or unique interests to perturb our thinking. But we don’t have to do it that way. Another theme was capitalism, and how it is or isn’t intertwined with democracy. How would we govern ourselves?

Government 193

Government Artificial Intelligence Technology Hacking 193

The Last Watchdog

JULY 26, 2023

July 26, 2023 — Protect AI , the artificial intelligence (AI) and machine learning (ML) security company, today announced it has closed a $35M Series A round of funding. to help organizations protect ML systems and AI applications from unique security vulnerabilities, data breaches and emerging threats. Seattle, Wash.,

Software 188

Software Digital transformation Marketing Artificial Intelligence 188

Schneier on Security

JULY 21, 2023

Imagine a future in which AIs automatically interpret—and enforce—laws. In New York, AI systems equipped with facial recognition technology are being used by businesses to identify shoplifters. Ubiquitous AI-powered surveillance in society will be necessary to enable such automated enforcement.

Surveillance 215

Surveillance Retail Technology Big data 215

Schneier on Security

SEPTEMBER 11, 2023

Historically, major disasters have needed to occur to spur regulation—the types of disasters we would ideally foresee and avoid in today’s AI paradigm. AI and robotics companies don’t want this to happen. Today’s AI-related robot deaths are no different from the robot accidents of the past.

Artificial Intelligence 298

Artificial Intelligence Technology Risk Government 298

Schneier on Security

AUGUST 13, 2021

Researchers are using GPT-3 to attempt to solve that problem: The researchers used OpenAI’s GPT-3 platform in conjunction with other AI-as-a-service products focused on personality analysis to generate phishing emails tailored to their colleagues’ backgrounds and traits.

Phishing 327

Phishing Risk Social Engineering Artificial Intelligence 327

Security Boulevard

FEBRUARY 22, 2024

Data collection and use policies need to be reexamined because of AI. The post TEGWAR, AI and the FTC – Gov’t Agency Warns of Deceptive AI Contract Language appeared first on Security Boulevard. The FTC is trying to address the issue.

Data collection 67

Data collection Data privacy Risk Government 67

Schneier on Security

NOVEMBER 11, 2022

A Hacker’s Mind: How the Powerful Bend Society’s Rules, and How to Bend them Back isn’t about hacking computer systems; it’s about hacking more general economic, political, and social systems. Or norms, since the “rules” aren’t always formal. I have a new book coming out in February.

Artificial Intelligence 177

Artificial Intelligence Hacking Internet Internet 177

The Last Watchdog

DECEMBER 14, 2023

ai Antani Many speculated that the ransomware attack on a Toyota supply chain player in Kojima, Japan was in retaliation for Japan’s aid to Ukraine. Generative AI is expected to supercharge the velocity and precision of attacks. Generative AI is expected to supercharge the velocity and precision of attacks.

Cybersecurity 235

Cybersecurity Marketing Encryption CISO 235

Troy Hunt

JANUARY 5, 2023

The Twitter incident consumed a heap of my time before, during and after this live stream, but then I go and get a sudden itch to do stuff like the number plate capturing and, well, there goes even more hours I don't have. If you've got a footprint in the fediverse, please go and give the account a follow. See how passkeys work.

Password Management 230

Password Management Accountability Passwords 230

Malwarebytes

DECEMBER 26, 2023

But most phishers aren’t very good, and the success rate is relatively low: In 2021, the average click rate for a phishing campaign was 17.8%. However, now cybercriminals have AI to write their emails, which might well improve their phishing success rates. The email contains an attachment you weren’t expecting.

Phishing 120

Phishing Software Risk Cybersecurity 120

Malwarebytes

NOVEMBER 8, 2023

However, students that are using those tools shouldn’t be receiving grades that they don’t deserve. Several programs that can filter out artificial intelligence (AI) texts have been created and tests are ongoing, but the success rate of these, mostly AI-based tools, hasn’t been great.

Artificial Intelligence 130

Artificial Intelligence Internet Internet Software 130

The Last Watchdog

JUNE 26, 2023

Related: A call to regulate facial recognition Although AI has become a powerful weapon, there’s concern it might be too effective compared to human cybersecurity professionals — leading to layoffs and replacements. However, the truth is that automated AI tools work best in the hands of cybersecurity professionals instead of replacing them.

Cybersecurity 178

Cybersecurity Artificial Intelligence Cyber threats Cybercrime 178

Security Affairs

OCTOBER 23, 2023

The Philippine defense ordered its personnel to stop using AI-based applications to generate personal portraits. The Philippine defense warned of the risks of using AI-based applications to generate personal portraits and ordered its personnel to stop using them. issued the directive to ban the AI-based applications.

Identity Theft 116

Identity Theft Social Engineering Data collection Risk 116

Security Affairs

APRIL 14, 2024

Crooks manipulate GitHub’s search results to distribute malware BatBadBut flaw allowed an attacker to perform command injection on Windows Roku disclosed a new security breach impacting 576,000 accounts LastPass employee targeted via an audio deepfake call TA547 targets German organizations with Rhadamanthys malware CISA adds D-Link multiple (..)

Data breaches 102

Data breaches Social Engineering Malware Hacking 102

LRQA Nettitude Labs

FEBRUARY 29, 2024

BloreBank Chatbot is a prompt injection game where you try to trick the AI into giving away sensitive information. The backend AI draws from actual scenarios encountered in the field. The backend AI draws from actual scenarios encountered in the field. The main aim of this game is to raise awareness about AI system security.

Banking 82

Banking Artificial Intelligence Cybersecurity 82

Daniel Miessler

APRIL 2, 2023

There is much being said right now about the AI Alignment Problem. There’s more to it, but it’s essentially the question of how to keep AI from destroying us when it wakes up. We don’t know what AI will do when it wakes up, but we should try present it our best self. All languages. All implementations.

Information Security 168

Information Security 168