top of page


OpenAI’s Goblin Problem Reveals Hidden AI Training Risks
OpenAI’s latest AI models developed an unexpected habit: talking about goblins. After users and employees noticed ChatGPT and related tools using odd creature-based metaphors, OpenAI investigated why words like “goblin,” “gremlin,” and even references to raccoons, trolls, ogres, and pigeons were appearing more often in responses. What started as a strange but mostly harmless language quirk became a useful example of how AI training can accidentally reward and spread unusual b
Covertly AI
Apr 303 min read


Elon Musk and Sam Altman Face Off in OpenAI Trial
Elon Musk and Sam Altman are heading into one of the most closely watched legal battles in the artificial intelligence world, as Musk’s long-running lawsuit against OpenAI finally moves to trial. The case begins with jury selection in Oakland, California, before U.S. District Judge Yvonne Gonzalez Rogers, with opening arguments expected shortly after. Musk, Altman, OpenAI president Greg Brockman, and Microsoft CEO Satya Nadella are all listed among the possible witnesses. At
Covertly AI
Apr 273 min read


Lovable.dev Security Flaw Exposes Risks in AI-Built Apps
Lovable.dev is under growing pressure after multiple reports described a serious security flaw that may have exposed thousands of older projects to anyone with a free account. The issue reportedly affected projects created before a November 2025 cutoff and allowed unauthorized users to access full source code, database credentials, AI chat histories, and even live customer data. Researchers said the flaw was a Broken Object Level Authorization problem, meaning the platform c
Covertly AI
Apr 204 min read


OpenAI Trained Its AI on Workplace Data: Should Employees Be Worried?
OpenAI is reportedly asking third-party contractors to upload real work they have completed in past and current jobs as part of an effort to improve the quality of training data used for its AI models. According to reporting from Wired, OpenAI is working with a training data company called Handshake AI to collect authentic workplace materials created by contractors, rather than relying solely on synthetic or hypothetical examples. This initiative appears to reflect a broade
Covertly AI
Jan 123 min read


Wisconsin’s AI Data Boom Sparks Environmental and Energy Debate
Artificial intelligence has driven a wave of massive infrastructure projects across Wisconsin, where major tech firms are investing billions to build sprawling data centers. These facilities, which power AI models and cloud computing, have become a symbol of both technological progress and growing concern. Supporters see them as engines of economic growth, while environmentalists and residents warn that their enormous energy and water demands threaten the state’s resources an
Covertly AI
Nov 6, 20254 min read


Apple Partners with Google Gemini to Run Siri with AI Power
Apple is preparing a major overhaul of its Siri voice assistant, and to get there, it’s turning to an unexpected ally, Google. According to Bloomberg and Reuters, Apple plans to use Google’s 1.2 trillion-parameter Gemini artificial intelligence model to power a reimagined version of Siri, marking one of the tech giant’s most significant partnerships in recent years (Reuters; Bloomberg; Financial Post). The deal, which is expected to cost Apple about $1 billion per year, is de
Covertly AI
Nov 6, 20254 min read


Microsoft's Vision for American AI: A Foundation for Global Leadership and Prosperity
Microsoft has unveiled an ambitious vision to position the United States as a global leader in artificial intelligence (AI) by leveraging...
Covertly AI
Jan 8, 20252 min read


Wordware Raises $30 Million to Transform AI Development into Natural Language Creation
Worldwide, a San Francisco-based startup is redefining how artificial intelligence is developed by making it as simple as writing in a...
Covertly AI
Nov 22, 20243 min read
bottom of page
.png)

