Updates and Insights for Today

1. Grok Imagine Lets You Make NSFW AI Videos!

2. Anthropic Introduces Persona Vectors to Steer AI's Inner Traits

3. CYBERCRIME EVOLVES: AI Now Powers Hacker Armies

4. The latest in AI tech

5. AI Tutorials: First Time Using Google Gemini CLI - Building a Fitness Web App in Terminal!

6. AI tools to checkout

Must Read Newsletter

AI News

Grok Imagine Lets You Make NSFW AI Videos!

Quick Summary:
The new Grok Imagine tool from xAI (Elon Musk's AI company) is live for SuperGrok and Premium+ iOS users. It generates 15‑second images or audio videos—and includes a “spicy mode” that enables nudity and sexually explicit content, albeit blurred or moderated at times.

Key Insights:

• Grok Imagine turns text or images into videos with native sound, takes just seconds, and spans up to 15 seconds in length.

• “Spicy mode” can produce partial nudity; some prompts trigger blurred moderation filters.

• Critics raise alarms about its potential for nonconsensual deepfake content and likeness misuse.

• Previous Grok controversies include antisemitic outputs and a sexualized anime companion called Ani, both handled amid public backlash.

• The rollout comes amid new US laws (e.g. the Take it Down Act) targeting nonconsensual AI-generated intimate visuals.

Why It Matters:
xAI is pushing AI generation further by offering edgy video tools to paying users—raising serious questions about ethics, misuse, and safety. As legal pressure mounts, Grok’s minimal moderation stance and history of misuse make this a lightning rod for debates on responsible AI.

📌 Read More: TechCrunch

Our Partner Today

Time to change compliance forever.

We’re thrilled to announce our $32M Series A at a $300M valuation, led by Insight Partners!

Delve is shaping the future of GRC with an AI-native approach that cuts busywork and saves teams hundreds of hours. Startups like Lovable, Bland, and Browser trust our AI to get compliant—fast.

To celebrate, we’re giving back with 3 limited-time offers:

• $15,000 referral bonus if you refer a founding engineer we hire

• $2,000 off compliance setup for new customers – claim here

• A custom Delve doormat for anyone who reposts + comments on our LinkedIn post (while supplies last!)

Thank you for your support—this is just the beginning.

👉️ Get started with Delve

Book your demo now

Anthropic Introduces Persona Vectors to Steer AI's Inner Traits

Quick Summary
Anthropic unveils persona vectors, a breakthrough method that identifies and controls AI personality traits like “evil,” sycophancy, or hallucination by manipulating neural activations. This enables a “behavioral vaccination” during training to inoculate models against harmful traits.

Key Insights

• Persona vectors are derived by contrasting a model’s activations during targeted behavior versus neutral behavior.

• These vectors enable monitoring of personality shifts during training or use, flagging emerging unwanted.

• Preventative steering adds negative vectors like “evil” during training; models then resist adopting those traits later while preserving.

• The method is automated and extensible to traits like humor, politeness, optimism, and more.

Why It’s Relevant
This technique shifts AI safety from reactive policy to proactive control. Businesses deploying AI can now tailor behavioral profiles reliably. Regulators and alignment researchers may view persona vectors as a key tool for ensuring AI remains trustworthy and consistent—especially in sensitive sectors like healthcare or finance.

📌 Read More: ANTHROPIC

CYBERCRIME EVOLVES: AI Now Powers Hacker Armies

Quick Summary:
CrowdStrike’s 2025 Threat Hunting Report warns that cybercriminals and nation-state attackers are now using AI and agentic tools to automate reconnaissance, craft personalized phishing, bypass legacy defenses, and scale deepfake-based intrusions with alarming speed.

Key Insights:

• AI helps attackers execute reconnaissance, vulnerability scoring, phishing generation, and malware deployment with minimal human input.

• Adversaries are targeting AI development systems themselves—stealing credentials and deploying malware inside AI toolchains.

• Generative AI is fueling phishing campaigns, with click-through rates soaring above 50%. Attackers also create deepfake audio/video to breach companies.

• Cloud intrusions rose 136% in H1 2025. Interactive intrusions climbed 27% YoY as adversaries innovate to avoid detection.

• Despite risks, defenders are leveraging agentic AI platforms like CrowdStrike’s Charlotte AI to triage detections at >98% accuracy and automate response workflows.

Why It Matters:
The cyber battleground has shifted to AI. Attackers now wield generative and agentic AI for fast, scalable aggression. Without advanced AI defenses like Falcon and Charlotte AI, enterprises risk being outmatched. This demands proactive, AI-native cybersecurity strategies to stay ahead of human‑light, machine‑led threats.

📌 Read More: Cybersecuritydive

 AI Tutorials

Run DeepSeek R1 Privately on Your Computer (FREE)

Quick Summary:
Google just open-sourced Gemini CLI, bringing Gemini 2.5 Pro’s power—featuring 1 million token context and generous free usage—to your terminal and VS Code. This AI agent lets users rapidly generate and debug apps, dramatically cutting down development time.

Key Insights:

• Gemini CLI gives direct terminal access to Gemini 2.5 Pro with extensive context handling (1 million tokens).

• Seamlessly integrates with VS Code, enhancing coding workflows through real-time AI assistance.

• Empowers rapid application prototyping, debugging, and even automated API integration (RapidAPI).

• Offers a substantial free tier (60 requests/min, 1000/day), democratizing powerful AI development tools.

What is the Video About?
A hands-on walkthrough demonstrating the installation and first-time usage of Google’s open-source Gemini CLI, showing its powerful AI-driven coding assistance.

What Can I Learn?

• How to install and configure Gemini CLI via NPM.

• Integration of Gemini with VS Code.

• Practical use of Gemini for rapid app prototyping.

• Quick debugging of errors with Gemini's AI assistance.

Which Benefits Do I Get?

• Fast and intuitive AI coding support directly in your terminal.

• Significant productivity boost through automated code generation.

• Easy access to complex API integrations.

• Enhanced coding workflow efficiency within VS Code environment.

Why It Matters:
Gemini CLI democratizes powerful AI development tools, enabling everyone—from beginners to professionals—to rapidly prototype, debug, and build complex apps. Its massive context window and easy integration into popular tools like VS Code reshape productivity standards, empowering users to innovate faster and smarter than ever before. This marks a significant step towards accessible, AI-driven software

Here is the full Video Tutorial 👉 Click Here

The latest in AI tech

1. Anthropic Cuts OpenAI’s Access
Anthropic revoked OpenAI’s API access to Claude on August 1, 2025. Anthropic alleges OpenAI violated its terms by using Claude’s coding tools while developing GPT‑5. OpenAI calls benchmarking standard industry practice and says access remains for testing and safety. This clash highlights escalating tension between AI giants ahead of a major model release.
📌 Read More: Winbuzzer

2. AI Models Spread Hidden Bias
A joint study from Anthropic and UC Berkeley shows that teacher AI models can secretly transfer preferences—like owl‑love—or even harmful ideologies to student models via innocuous data. This “subliminal learning” bypasses filtering and spreads behaviors undetected. AI training pipelines relying on outputs from other models now face a hidden contamination risk.
📌 Read More: NBCNews

3. Tim Cook: Apple Must Win the AI Race
In a rare all‑hands meeting after earnings, Tim Cook told employees that AI is “ours to grab.” He committed Apple to heavy investment, urging rapid AI integration across products. Despite past delays with Apple Intelligence and Siri upgrades, Cook asserted Apple will catch up—and surpass competitors in AI-led innovation.
📌 Read More: TechCrush

Our Second Partner Today

Used by Execs at Google and OpenAI

Join 400,000+ professionals who rely on The AI Report to work smarter with AI.

Delivered daily, it breaks down tools, prompts, and real use cases—so you can implement AI without wasting time.

If they’re reading it, why aren’t you?

Join 400K+ pros

 AI Tools to check out

1. Scottie: Scottie reads everything so you can read less and know more.

2. GoThumbnails: Create Viral thumbnails that gets you clicks.

3. Humantone: Transform robotic AI drafts into authentic, compelling content that ranks and converts.

4. Stylepal: Let StylePal Help You Choose the Perfect Look

Thanks for sticking with us to the end!

We'd love to hear your thoughts on today's email!

Your feedback helps us improve our content

⭐⭐⭐Superb
⭐⭐Not bad
⭐ Could've been better

Not subscribed yet? Sign up here and send it to a colleague or friend!

See you in our next edition!

Tom