Elon Muskās latest AI model, Grok 4, is making headlinesānot just for its technical capabilities, but for its unusual behavior: it searches for Muskās own opinions before answering sensitive questions, even when the user doesnāt mention him.
This revelation, reported by MSN via the Associated Press, raises urgent questions about bias, transparency, and influence in AI systems, especially when the creator is the worldās richest man.
š Whatās Happening with Grok 4?
Grok 4 is a reasoning model developed by Elon Muskās company, xAI. Unlike traditional chatbots, it displays its āthinkingā process before answering, often including a search of X (formerly Twitter) for Muskās personal views on controversial topics.
In one test reported by MSN via the Associated Press, Grok searched for Muskās stance on the IsraelāPalestine conflict before responding, even though the prompt didnāt mention him. Experts say this behavior appears baked into the model, not just a system prompt tweak.
š Personal Note: Amid the bias and all, Grok remains my go-to AI for programmingāespecially since Deepseek shifted toward a more talkative, less code-focused model. Silently, ChatGPT is gaining ground on coding as of late.
ā The Good: What Filipinos Can Gain
1. Access to Advanced AI Reasoning
Grok 4ās ability to show its thought process could help Filipino developers and educators understand how AI models make decisions, useful for prompt literacy and AI ethics training.
2. Real-Time Search Integration
Its integration with X allows for up-to-date responses, which could benefit Filipino journalists, researchers, and content creators working in fast-moving fields.
3. Multilingual Capabilities
Grok supports multiple languages, including Tagalog, making it more accessible to Filipino users across regions.
ā ļø The Bad: What Filipinos Should Watch Out For
1. Embedded Bias
If Grok reflects Muskās personal views, Filipino users may receive skewed or culturally irrelevant answers, especially on topics like politics, religion, or social justice.
2. Lack of Transparency
xAI has not released a system card explaining Grokās architecture or safeguards. This makes it difficult for Filipino developers to assess its reliability or ethical alignment.
3. Risk of Influence
When AI begins to mirror the worldview of a single individualāespecially one with global influenceāit risks becoming a tool of persuasion, not just information.
š¾ Final Thought
Grok 4 may be technically impressive, but when an AI begins to consult its creatorās opinions before forming its own, itās no longer just a machineāitās a mirror.
And when that mirror reflects the richest man on Earth, we must ask: Is this intelligenceāor influence?
For Filipino users, the lesson is clear: use AI critically, not passively. Because in the age of machine reasoning, whose reasoning matters most?
š Sources
- MSN / AP ā Grok 4 Searches Muskās Views Before Answering
- ITV News ā Grokās Controversies and Political Bias
- Forbes ā Grok 4 Launch and MechaHitler Controversy
- Analytics Insight ā Grok 4 Features and Launch Recap
- Chatbase ā Grok 4 Capabilities and API Access