For core business automations, agentic AIs that "guess" are expensive and unreliable. A superior approach uses tools that convert natural language into deterministic, code-like workflows, which run consistently and use AI only when necessary.

Fully autonomous agents are not yet reliable for complex production use cases because accuracy collapses when chaining multiple probabilistic steps. Zapier's CEO recommends a hybrid "agentic workflow" approach: embed a single, decisive agent within an otherwise deterministic, structured workflow to ensure reliability while still leveraging LLM intelligence.

Despite advancing capabilities, AI models like ChatGPT can exhibit surprising fragility. They can get stuck in nonsensical loops or "spiral out" on straightforward queries, such as questions about Zapier integrations. This unpredictable fallibility demonstrates that model reliability remains a significant challenge, eroding user trust for critical tasks.

Wade Foster argues that for most business automation, delegating tasks to an AI agent is inefficient. Instead, an AI should be used to *build* a deterministic workflow (i.e., traditional code) which is cheaper and more reliable for execution. AI's reasoning is best reserved only for parts that truly require it.

According to Zapier's CEO, the optimal approach is a hybrid one. AI agents excel at interpreting natural language to construct complex workflows. However, the execution of these workflows should rely on deterministic, predictable code for reliability and lower cost, since 80% of agent tasks don't need AI's judgment.

For robust personal automation, pure agentic loops are inefficient and unreliable. Zapier's CEO structures his agents to use deterministic code for predictable tasks like fetching calendar events, which is cheaper and more reliable. AI reasoning is reserved only for specific steps like summarization, creating a powerful hybrid.

The most significant risk from AI agents currently isn't sophisticated prompt injections but simple misinterpretations of instructions that lead to 'unintended actions.' This makes focusing on controlling outcomes more effective than trying to identify the source of a faulty instruction, be it a hallucination or an attack.

A key asymmetry exists in AI deployment: it has become much easier to use AI to generate exact, predictable automation software (design time). However, using probabilistic AI agents to directly execute enterprise processes (run time) remains just as difficult and ungovernable as before.

Fully autonomous AI agents are not yet viable in enterprises. Alloy Automation builds "semi-deterministic" agents that combine AI's reasoning with deterministic workflows, escalating to a human when confidence is low to ensure safety and compliance.