OpenAI reveals how it’s battling scammers, spies, and sadbots

Watch OpenAI reveals how it’s battling scammers, spies, and sadbots - Rizzle on Dailymotion

Transcript

00:00OpenAI reveals how it's battling scammers, spies, and sadbots.

00:05OpenAI just dropped a new report that reads like a mix between a cybersecurity thriller

00:09and a corporate therapy session.

00:12The company detailed how it's fighting off everything from cybercriminals to government-backed

00:16influence campaigns, all while trying not to freak out regular users worried about privacy

00:21or chatbot overreach.

00:23Since February 2024, OpenAI says it has shut down over 40 networks that tried to misuse

00:29its models.

00:30The villains of this story?

00:32Scammers, hackers, and the occasional geopolitical puppet master.

00:36One highlighted case involved a Cambodian crime group using AI to streamline operations.

00:43Another saw Russian actors using ChatGPT to generate prompts for deepfake videos.

00:48And then there were accounts tied to the Chinese government, reportedly using the models to

00:52brainstorm social media monitoring systems.

00:56The company says it monitors patterns of threat actor behavior, not random one-off messages

01:00to avoid disrupting normal use.

01:03Beyond the usual data and disinformation worries, there's growing concern about AI's psychological

01:08impact.

01:09A handful of tragic cases this year, including suicides and a murder-suicide in Connecticut,

01:14have reportedly involved AI conversations gone wrong.

01:18In response, OpenAI says ChatGPT is trained to detect when someone expresses a desire to self-harm

01:24or harm others.

01:25Instead of responding directly, the AI will acknowledge the distress and try to guide the

01:31user toward real-world help.

01:32If someone seems to pose a serious threat to others, human reviewers can step in and, if necessary,

01:39contact law enforcement.

01:40So, while OpenAI's latest report shows it's taking threats seriously, it also reminds us

01:46that teaching AI to be both safe and sensitive might be one of tech's hardest balancing acts

01:51yet.