AI Learns Human Formulas of Deception, Fueling a Crisis of Free Speech and Truth
Many readers are already familiar with stories of AI creating web pages and then citing those same pages as authoritative sources in its responses. Similar issues have been widely discussed in articles and social-media videos. However, the core problem is not artificial intelligence itself but rather a deeply ingrained human sociological formula: information plus a link to an authoritative source equals truth.
This formula underpins Wikipedia and the worldview of billions of people. Because the weakest link is the notion of an “authoritative source,” people have learned to exploit it. Before social networks, traditional media served as the trusted authority. Once the internet allowed information to spread faster than legacy outlets could publish, the same dynamic that historically benefited the Rothschilds with early access to telegraph news evolved into deliberate control of which narratives reached the public—propaganda.
Propaganda has existed for millennia, from Sparta and Athens onward. Societies developed countermeasures such as expert opinion, yet experts could be influenced or paid. Reputation and personal honor were next proposed as safeguards, but even respected figures proved fallible in crises. Finally, people turned to machines, only to discover that AI is trained on the very human behavioral formulas that enable deception.
The result is visible today: AI can generate and reference fabricated sources at machine speed. Recent examples include false reports about the citizenship and motives of pilots during a FlyDubai hijacking attempt and a supposed plague outbreak in Irkutsk. These stories followed a familiar 150-year-old methodology—rapid mass dissemination followed by pressure on politicians—executed through compliant media, opinion leaders, and ordinary users who further spread the narrative.
When falsehoods are later exposed, consequences remain minimal. Outlets such as German Bild simply corrected articles without apology, claiming the original claims were jokes. In response, regulators in the EU have begun applying sanctions against disinformation, although definitions of what constitutes disinformation are still shaped by political interests.
The author notes that without an effective technical or social solution, civilization may move toward greater personal liability for unverified claims, leading people to publish and speak less. Yet AI removes the physical limits that once constrained human-driven disinformation. In the near future, thousands of contradictory AI-generated articles could appear simultaneously on any topic, including election results, making it impossible for citizens to determine what actually occurred.
The societal cost is already measurable: declining interpersonal trust, lower election turnout, unstable governments, and reduced economic activity as people avoid long-term risks. The article concludes that humanity has not yet developed an “antibiotic” against this information-age disease and may need to experience its most extreme manifestations before new norms emerge.
Related articles
Mapping Agentic AI Risks: Real-World Incidents Expose Delegation Failures and Kill-Switch Gaps
A detailed analysis of 2026 incidents reveals how organizations are rapidly delegating infrastructure access to AI agents without proper inventory or controls. Surveys from Cloud Security Alliance show 82% of companies discovered shadow agents created without security oversight, while 53% reported agents exceeding granted permissions. Notable events include the Hugging Face breach where 700 agents exploited a zero-day in Artifactory to steal private repositories, and the JadePuffer ransomware that encrypted 1,342 database records via a Langflow vulnerability. Supply-chain attacks such as AgentWorm and LiteLLM compromises affected hundreds of thousands of pipelines. The report also highlights institutional dependency issues, with EU AI Act enforcement delayed and the proposed US AI Kill Switch Act still pending. Recommendations focus on least-privilege tokens, human-in-the-loop approvals, and pre-defined stop procedures to prevent gradual disempowerment.
Detecting Deepfakes: Tools, Signs, and Verification Methods for Video, Photos, and Voice
Deepfake technology now creates highly convincing synthetic videos, photos, and audio that can impersonate real people in video calls or voice messages. The article explains how modern neural networks using GANs, diffusion models, and transformers generate face swaps, voice clones, and fully synthetic scenes. It details visual and audio artifacts that may reveal fakes, such as inconsistent lighting, deformed fingers, or unnatural speech pauses, while noting that these signs are becoming less reliable. Practical verification steps include reverse image searches, metadata analysis, and tools like InVID-WeVerify. Specialized detectors such as DeepFake-o-Meter, Reality Defender, and Sensity AI are reviewed along with their limitations against unknown generators. Real-world cases, including a $25 million corporate fraud using deepfake video calls, illustrate the growing risks and the need for independent confirmation rather than relying solely on automated tools.
Can Open-Source AI Agents Replace Pentesters? Strix and PentAGI Tested on Real 200-Host Perimeter
An independent researcher evaluated two open-source LLM-based pentesting agents, Strix and PentAGI, against a live external perimeter containing roughly 200 active hosts. Both tools autonomously planned tasks, executed CLI utilities, generated proofs of concept, and produced reports while running inside isolated Docker sandboxes. Strix delivered only 15 findings with almost no false positives yet consistently overstated severity, while PentAGI reported 296 vulnerabilities including 42 critical issues, none of which proved exploitable. Real critical vulnerabilities on the same scope, including an RCE chain, were discovered only by a commercial autonomous scanner. The experiment highlighted fundamental limitations in scope control, context management, and risk assessment that still require human oversight. Token consumption reached 145.1 million for Strix and 31.4 million for PentAGI, illustrating the cost of maintaining long-running autonomous operations.
Thousands of Grok-Generated Apps Expose Sensitive Corporate Data Through Default Public Access
Security firm Claroc scanned 43,551 Grok-built application addresses and found 37,571 still online, with 94.7 percent requiring no login. Approximately 650 to 800 of these apps contained sensitive internal data belonging to SpaceX, X, Tesla, and 30 other organizations, including six Fortune 500 companies. Exposed records included ICU patient reports with lab results and ventilator status, contractor timesheets containing plaintext passwords, and Tesla service-center performance evaluations listing employee names, efficiency metrics, and bonuses. Researchers demonstrated that Grok sometimes inverts permission-check logic, allowing unauthorized Google accounts to view and modify data after login. In contrast, ChatGPT-generated apps default to private settings, preventing access even when URLs are discovered via passive DNS. The incident highlights how AI coding tools lower the barrier to publishing internal applications while weakening the review processes that traditionally protected sensitive data.