Anish Acharya: Personal Agents Will Be Invisible
Anish Acharya, a general partner at Andreessen Horowitz, discussed the explosion of personal AI agents with David Pawlan, creator of Assistant Benchmark. Pawlan tested dozens of assistants on real tasks, covering email, booking, and financial administration. Both believe the most useful agents will become increasingly invisible, proactively checking in, seeking refunds, filing expenses, and handling daily minor tasks.
Pawlan's Assistant Bench compares assistants using the same prompts, such as booking a flight to Chicago or finding a vegan restaurant within five blocks, assessing success and follow-up on about 16 dimensions. He cataloged 122 assistants over a month, personally testing 26, and noted that many consumer products are merely rebranded generic agents. The discussion also covered how saving money tasks are superior to saving time tasks, Meta's Muse data and appeal, silent agents in group chats, the importance of proactivity as a moat, definitions of assistants and agents, and how Amazon intercepts Muse, Shopify opens up, along with an operational cost of about $20 per day for agents.
Acharya bets that the subsidy model is unstable and is willing to pay a high monthly fee for narrow, effective agents, which may be more robust than free generalists chasing millions of downloads. Interface candidates include apps, SMS, voice, and wearables; how much autonomy to grant and how agents place orders will shape the next layer of the internet. The episode was recorded on September 24, lasting about 51 minutes.
On the same day, the product showcase featured enterprise agents entering market shelves, car systems, and HR systems; this conversation shifted the focus back to personal administration. A daily cost of $20 implies that free generalists will rely on platform subsidies, while vertical expense reporting and flight price differences may encourage users to pay out of pocket.
In market mechanisms, what is purchased is an agent that can successfully complete real booking and reimbursement tasks, while what is sold is evaluation rankings and venture capital narratives. Funds enter the reasoning bill from consumer subscriptions or platform subsidies; beneficiaries are vertical agents that can connect to merchant interfaces, while those under pressure are shell agents that can only chat without changing flights. The event driver is the podcast launch, not a specific assistant winning.
The evaluation shifts from "like a person" to "was this ticket booked successfully." Being proactive without causing issues is closer to a moat than simply having another persona. Platform interception and openness determine whether an agent can place orders or only make recommendations.
Source: Public Information
ABAB AI Insight
a16z uses a review interview to set standards for the consumer agent sector: focusing on ticket bookings and reimbursements rather than parameters. Pawlan's catalog of 122 agents proves supply explosion with insufficient differentiation; testing 26 agents squeezes out the fluff into "pre-configured generic agents." The $20 daily fee is an economic hammer: if generalist assistants are to truly click web pages and make phone calls to complete tasks, reasoning and human backup will consume the free tier. Acharya reserves the thousand-dollar monthly fee for narrow scenarios, acknowledging that the winners in consumer agents may resemble tax software rather than super apps. Amazon intercepts Muse, while Shopify opens doors, indicating channel power precedes model power.
The capital path is that review sites provide scoring tools for investors, and funds incorporate these scores into project memos. Money-saving tasks have cash returns, and users are willing to authorize credit cards; time-saving tasks only have subjective value, leading to weaker willingness to pay. If proactive check-in results in a wrong flight booking, trust is zeroed out, making the moat "knowing when to ask for guidance." Silent agents in group chats shift the entry point from standalone apps to existing social threads, lowering opening costs and exposing errors to all group members.
Analogous examples include early price comparison plugins, Mint financial assistants, and Siri, which has stagnated on weather for a decade. The industry phase sees personal agents transitioning from chat personas to administrative execution, with interface openness determining their potential scale. Enterprise agents are already selling full-time equivalents; personal agents are still seeking payers for the $20 daily fee. The first to connect airline rebooking with bank refunds will leave the demo rankings.
Structural judgment indicates a shift in pricing power. The mechanism is that once task results are verifiable, fees transition from seat-based to a share of successfully completed tasks or subscriptions. This shift occurs because users will not pay for yet another chat box but will pay for recovered ticket price differences. If platforms close order interfaces, pricing power remains in app stores; if opened, pricing power shifts to agents capable of stable execution.
ABAB News · Cognitive Laws
Money-saving agents are easier to charge than time-saving agents.
Proactivity is a moat; chaotic proactivity leads to churn.
$20 a day forces narrow and expensive, pushing back free generalists.