AI-powered scams and what you can do about them

In recent years, AI technology has advanced significantly, making it easier and cheaper to create realistic synthetic media, which has unfortunately led to a rise in AI-powered scams. Voice cloning, for instance, can generate convincing fake versions of loved ones from just a few seconds of audio, leading to distressing scams where victims are tricked into sending money to scammers posing as family members in trouble. To protect yourself, it's crucial to treat any unsolicited messages or calls from unknown sources with suspicion and verify the person's identity through a separate, trusted method.

Another area where AI has enhanced scams is personalized phishing and spam. With data breaches being common, scammers can use text-generating AI to craft highly customized and believable emails that trick individuals into clicking malicious links or opening dangerous attachments. The best defense against this is vigilance—avoid clicking on links or opening attachments from unknown senders and always verify the authenticity of any suspicious emails with a second, knowledgeable person if needed.

AI-generated deepfakes pose a particularly frightening threat, enabling the creation of fake intimate images for blackmail without needing real compromising material. These images can be realistic enough to cause distress, but often lack distinguishing marks and may have other inconsistencies. If targeted by such scams, victims should report the incident to authorities as it's not just a scam but also harassment. Legal measures and growing awareness can help combat this disturbing use of AI technology.

Transform your Business with a Modern Client Portal built by AI

If you run a service business, it's crucial to up your game to survive the AI revolution and rising competition. Clients expect more than endless email threads from you. With Hubflo, you can

• elevate your client experience and streamline operations with a powerful client portal built by AI.

• Customize your portal with 10 built-in tools for file sharing, e-signing, approvals, messaging, tasks, billing, and more.

• Integrate with 500+ external apps like Calendly, Figma, Loom, Drive, and more.

Get started today and impress clients with a white-label portal!

Amazon hires founders away from AI startup Adept

Adept, an AI startup specializing in developing AI-powered agents for software-based tasks, has entered into a licensing agreement with Amazon, allowing the e-commerce giant to leverage Adept’s technology. This move sees Adept's co-founders, including CEO David Luan, Augustus Odena, Maxwell Nye, Erich Elsen, and Kelsey Szot, along with several other employees, transitioning to Amazon. Adept, however, is not shutting down; instead, Zach Brock, head of engineering, will step in as CEO, redirecting the company's focus toward agentic AI solutions. Despite discussions with Meta and Microsoft for potential acquisition, the licensing deal with Amazon provides Adept with a crucial lifeline.

Amazon stands to benefit significantly from this deal, gaining both valuable talent and technology to bolster its generative AI initiatives. Luan and his team will now work under Rohit Prasad, former head of Alexa, who is leading a new team focused on building large language models and practical AI solutions. This strategic acquisition aligns with Amazon's vision to develop digital agents capable of automating software workflows, accelerating their roadmap for creating sophisticated AI-driven solutions for both consumer and enterprise applications.

Founded two years ago, Adept had ambitious goals of creating an AI model capable of performing tasks across various software tools using natural language. Despite raising over $415 million and achieving a valuation of around $1 billion, the startup faced internal challenges, including the loss of two co-founders and delays in product development. The market for AI agents has become increasingly competitive, with startups like Orby and Emergence entering the fray. Adept’s partnership with Amazon may be the push needed to bring their vision to fruition, though it remains to be seen if the company can overcome its previous hurdles.

Microsoft’s AI boss thinks it’s perfectly OK to steal content if it’s on the open web

Microsoft AI boss Mustafa Suleyman recently made a controversial statement regarding the use of online content, suggesting that once content is published on the open web, it becomes "freeware" that anyone can freely copy and use. In an interview with CNBC’s Andrew Ross Sorkin, Suleyman claimed that the longstanding social contract since the '90s has treated open web content as fair use. However, this view contradicts established copyright laws, which automatically protect a creator's work upon creation, without the need for registration or any special licenses.

Suleyman's comments come amid multiple lawsuits against Microsoft and OpenAI, alleging the theft of copyrighted online stories for training generative AI models. Critics argue that fair use is a legal defense determined by courts, not a social agreement. The law requires careful consideration of what is copied, why, how much, and the potential impact on the copyright owner. Suleyman's interpretation seems to overlook these legal nuances, raising concerns about the ethical and legal boundaries of AI training practices.

Furthermore, Suleyman acknowledged the importance of the robots.txt file, which specifies which bots can scrape a website, suggesting it might deter content misuse. However, robots.txt is not a legal document but rather a guideline. Despite this, some AI companies, including Microsoft partner OpenAI, reportedly ignore these guidelines. This disregard for established norms and legal frameworks highlights the ongoing tensions and uncertainties surrounding AI and copyright law.

Related: Perplexity’s Grand Theft AI

Other stuff

• Apple’s Devices Are Lasting Longer, Making AI Strategy Even More

• MIT’s soft robotic system is designed to pack groceries 🔥

• Robot cats, dogs, and birds are being deployed amid an ‘epidemic of loneliness’ 🔥

• India’s biggest AI startups based on how much money they’ve raised

• Chrome is adding window​.ai — a Gemini Nano AI model right inside your browser. 🔥🔥

• Figma will use your content to train its AI

• 'Skeleton Key' attack unlocks the worst of AI, says Microsoft 🔥

• Quora’s Chatbot Platform Poe Allows Users to Download Paywalled Articles on Demand 🔥

• Bill Gates says not to worry about AI gobbling up energy, tech will adapt

• OpenAI Blocks China and Hong Kong Access

Hope you enjoyed today's newsletter

⚡️ Join over 200,000 people using the Superpower ChatGPT extension on Chrome and Firefox.

OR