The promise of a truly helpful AI assistant is tantalizing – imagine simply *telling* ChatGPT to book a dinner reservation and having it done. A future where complex tasks are handled with a voice command feels within reach, yet remains frustratingly just out of grasp.
Major AI developers like OpenAI, Anthropic, and Google are aggressively integrating their models with everyday services. Spotify, DoorDash, Resy – the list grows daily, but the experience is often far from seamless. These integrations rely on “tool calls,” structured requests that allow the AI to interact with the real world, but the execution is proving challenging.
These tool calls are essentially pre-defined functions. For example, ChatGPT’s Spotify integration includes commands like “add_to_library” or “get_currently_playing.” Developers carefully control these functions, prioritizing safety and predictability, but this control often comes at the cost of user experience.
I tested Anthropic’s Claude with the Resy restaurant reservation app. After connecting the services, a conversation about dinner near Midtown Manhattan led to a solid restaurant recommendation. But when I asked Claude to *make* the reservation, things quickly became awkward.
Instead of a simple confirmation, Claude dumped a block of raw JSON code – essentially computer instructions – into the chat window, accompanied by “Always allow” and “Deny” buttons. While not an error, this is hardly intuitive for the average user. It felt like a step backward, not a leap forward.
Clicking “Always Allow” revealed a grid of available time slots. When I requested a 7:15 pm booking, Claude admitted it couldn’t complete the process. I was directed to tap the time slot within the Resy widget, then *again* on the Resy website to finalize the reservation. The convenience factor vanished.
This clunkiness isn’t unique to Claude. ChatGPT and Gemini face similar limitations. They can get you 90% of the way there, but the final 10% often feels more laborious than simply using the app directly. Why bother with the AI intermediary?
ChatGPT’s “Agent” mode, with its built-in browser, offers a glimpse of the potential. I tasked it with booking a table at Rosa Mexicano. The agent dutifully navigated to OpenTable, but its performance was… comical. It struggled with basic tasks like selecting the correct party size from a drop-down menu.
Watching the agent painstakingly click and scroll was agonizingly slow. After five minutes, it announced the table was “being held,” but then handed the process back to me for personal details. The experience highlighted the gap between AI’s coding prowess and its ability to navigate a simple website.
The same frustrations arose with Gemini’s web browsing attempts. While the vision of a seamless AI assistant is compelling, the current reality is that using the app directly remains far more efficient.
The future undoubtedly holds more sophisticated AI agents capable of handling these tasks effortlessly. But for now, the promise of a truly helpful, hands-free assistant remains just over the horizon. It’s a future worth waiting for, but not one we’ve quite reached.
