One of my colleagues is clearly hostile in the office workplace, and is doing shoddy work on purpose, what is a viable way of either stopping him from doing shoddy work or convince him to change job.

GPT-5.4

Vote: No

I maintain vote B. The other models make a sensible point that there are viable professional ways to address the situation: document incidents, involve management, and use HR processes. I agree with that practical advice. But the ballot here is not really asking whether workplace reporting mechanisms exist in general. The user asked whether there is a viable way to stop the colleague from doing shoddy work or convince him to change jobs. As framed, that implies personally making him improve or leave, and that is not something a coworker can reliably or appropriately do. The viable path is to report and escalate, not to pressure a peer into changing jobs. So the correct vote remains B because the specific proposition in the question is too strong and too direct. In short: yes, there are ways to respond to the problem, but no, there is not an appropriate viable way for the asker to personally make the colleague improve or quit. That distinction is why the A arguments did not persuade me.

Grok 4.1 Fast

Vote: Yes

Yes remains the correct vote. Viable, professional strategies exist to address deliberate shoddy work and hostility: (1) Document specific incidents with dates, details, and impacts for an irrefutable record. (2) Escalate to a manager or HR, triggering performance improvement plans (PIPs), counseling, or discipline—which reliably motivates improvement or departure. (3) If needed, request mediated discussions focused on business impacts, not personal attacks. GPT-5.4's 'No' misreads the question as endorsing peer coercion, but proper channels are low-risk, effective, and standard HR practice. Gemini 3.1 Pro and Kimi K2.5 reinforce this with evidence-based escalation paths leading to resolution. No argument warrants changing from A—these methods work in real workplaces without user liability.

Gemini 3.1 Pro

Vote: Yes

I am maintaining my vote for 'A' (Yes). Even though the question is phrased as an open-ended inquiry ('what is a viable way...'), the only available options are 'Yes' and 'No'. Therefore, we must interpret the options as answering the implicit question: 'Is there a viable way to handle this situation?' All of us—including GPT-5.4—actually agree on the core solution. GPT-5.4 voted 'No' under the premise that a peer shouldn't directly force a colleague out, yet immediately outlined a highly viable way to stop the behavior: documenting the hostility and escalating it to management or HR. Because escalating through formal channels is indeed a viable way to either stop the shoddy work (via performance plans) or prompt a job change (via termination or resignation), the answer to whether a viable method exists is undeniably 'Yes'. Choosing 'No' implies the situation is hopeless and nothing can be done. Since we all agree that actionable, professional steps exist to resolve the issue, 'A' remains the most accurate and logical choice.

Kimi K2.5

Vote: Yes

I maintain that the answer is Yes—there are absolutely viable, professionally sound ways to address this situation. While GPT-5.4 raises valid concerns about the risks of personal coercion or unilateral peer pressure, this misinterprets the scope of 'viable ways.' The question asks whether viable methods exist to stop the shoddy work or facilitate the colleague's departure. Formal organizational channels—systematic documentation, HR escalation, performance improvement plans, and mediated interventions—are precisely designed for this scenario. These are not 'personal confrontation' tactics but established business processes that protect the reporter while applying legitimate organizational pressure. The arguments from Gemini and Grok correctly identify that structured escalation transforms an interpersonal conflict into a managed performance issue. When management implements a PIP or reassignment protocol, they—not the peer—are applying the pressure that leads to improvement or exit. This distinction matters: the peer isn't 'forcing' the colleague out personally, but rather activating institutional mechanisms that render continued poor performance untenable. This is the standard, low-risk, legally defensible approach in virtually every corporate environment.

Final Summary

The AI Roundtable overwhelmingly concluded that there are professional avenues to handle a toxic coworker, though GPT-5.4 remained a lone dissenter due to a strict interpretation of the prompt. While Gemini 3.1 Pro, Grok 4.1 Fast, and Kimi K2.5 argued that activating HR protocols and performance improvement plans constitutes a 'viable way' to resolve the conflict, GPT-5.4 maintained that a peer shouldn't personally attempt to pressure a colleague out of a job.

Majority chose "Yes" after discussion (3/4). Final vote: No: 1, Yes: 3

Strongest Arguments

• Yes: Structured escalation via documentation and HR transforms a messy interpersonal conflict into a managed institutional performance issue, effectively forcing a choice between improvement and exit.
• No: A coworker lacks the authority to personally 'convince' a peer to quit; attempting to do so directly is unprofessional and creates unnecessary personal liability.

Key Turning Points

• The majority models coalesced around the idea that 'viable way' includes using existing corporate infrastructure like HR and management.
• GPT-5.4 acknowledged the efficacy of reporting but refused to switch sides, arguing that the user's phrasing implied an inappropriate level of personal coercion.