ChatGPT Tasks: Scheduled Reminders at the 2025 Launch

Matthew Berman explains the initial beta of scheduled and recurring ChatGPT tasks, and places it in the context of AI assistants.

MBMatthew Berman연사 · Forward Future
채널
Matthew Berman
게시일
2025년 1월 14일
영상
2:57
읽기
2분
슬라이드
1
원문
영어
YouTube에서 열기
스크롤하면 영상이 오른쪽 아래로 작아져 계속 재생됩니다. 자막은 읽기 언어를 따릅니다.
읽기 언어2개 언어로 읽을 수 있음
이 글은 아직 한국어 번역이 없어 영어 원문으로 표시합니다.
요약

30초 요약

In January 2025, Matthew Berman describes the announced beta of ChatGPT Tasks, which lets users schedule one-time or recurring requests. He had not yet received access, so his account relies on launch reporting rather than a hands-on test. Availability, interface, and limits described in the video are historical and may have changed.

  1. 1Scheduling a requestBerman describes asking ChatGPT to follow up at a specified future time or on a recurring schedule.
  2. 2A launch report, not a testHe says the beta was rolling out to paid subscribers and that he did not yet have access.
  3. 3Suggested tasksThe contemporary report he cites says users must approve suggested tasks before creation.

핵심 정리

  1. Tasks were introduced as a way to schedule a request or reminder for a future time.
  2. The initial beta was described as supporting recurring requests and user-approved suggestions.
  3. Berman sees scheduling as a step toward assistants that act over time, while noting that scheduling itself is a familiar software capability.
전문

챕터별로 다듬은 전문 읽기

원본 자막만 보기
01

What Tasks was designed to do

The beta introduced scheduled reminders and requests in ChatGPT.

In January 2025, Matthew Berman reports that OpenAI is introducing Tasks in ChatGPT. The feature is intended to let a user describe something to do or remember and specify when to follow up. He sees it as a useful step toward a digital assistant, while noting that scheduling is a straightforward capability in conventional software.

Berman says the initial beta was rolling out to Plus, Team, and Pro subscribers, but he had not yet received access. His description is therefore based on launch reporting rather than his own test. The interface, eligible plans, limits, and product capabilities may have changed since this video.

02

One-time and recurring requests

Berman gives a daily weather report as an example and describes the launch interface.

Launch examples for scheduled tasks
슬라이드 1 / 1Launch examples for scheduled tasksContemporary reporting illustrates one-time reminders and recurring requests.

A request could be scheduled once or repeatedly: for example, a daily weather report at 7 a.m. Berman also describes an interface for viewing saved reminders and the model selector used at launch. Those specific navigation instructions are a snapshot of the original beta, not necessarily current instructions.

03

Suggestions and future assistants

He discusses approval for suggested tasks and the potential for more agent-like features.

The report he reads says ChatGPT could suggest tasks based on a conversation, with user approval required before creating them. Berman finds that useful for recurring information or future follow-ups. He reports an initial limit of ten active tasks and no announcement at the time about a free-tier rollout.

Berman connects scheduled tasks to the broader idea of agents that carry out work on a user's behalf, and looks ahead to browser-controlling systems. He stresses that a simple scheduled request can also be built with traditional scripting; his interest is in bringing the experience into a conversational assistant.

같은 채널의 다른 글

슬라이드 6장25:50EN
MBMatthew Berman· 2026년 10월 24일Kimi K3.1 Pro Tested: Strong 3D Demos, Uneven Results and Competitive PricingMatthew Berman · Forward FutureMatthew Berman tests the open-weight model on coding, 3D rendering and visual explanations, then compares benchmark results and prices.Open-weight modelsBenchmarksAI cost & performance+1
슬라이드 2장32:20EN
MBMatthew Berman· 2026년 10월 23일Who Is Local AI For? The Costs, Trade-offs and a Hybrid FutureMatthew Berman · Forward FutureMatthew Berman weighs data control and independence against hardware cost, setup complexity and inference speed.Local inferenceOpen-weight modelsAI cost & performance+1
45:08EN
MBMatthew Berman· 2026년 10월 21일Sam Altman on AI Discovery, Consciousness, Compute and a Human-Centered FutureMatthew Berman, Sam Altman · Forward FutureIn a conversation with Matthew Berman, Altman discusses scientific progress, personal assistants, recursive self-improvement and the importance of human agency.AI for sciencePersonal agentsConsciousness+2
0:32EN
MBMatthew Berman· 2026년 10월 21일Sam Altman on the Continuing Demand for Software WorkSam Altman, Matthew Berman · OpenAIIn a short interview excerpt, Altman says demand for software may be larger than expected even as AI changes how it is made.Future of WorkSoftware engineering코딩 에이전트
슬라이드 5장12:14EN
MBMatthew Berman· 2026년 10월 14일Gemini 4 Argon Tested: 3D Demos, a Spreadsheet Prototype and Benchmark ResultsMatthew Berman · Forward FutureMatthew Berman tests Google’s model on games and visual projects, then compares reported coding and agent benchmarks.GeminiBenchmarks코딩 에이전트+1
슬라이드 5장10:36EN
MBMatthew Berman· 2026년 10월 13일Meta Muse Wonder Tested: Strong Benchmarks, Mixed Demos and Quota FrictionMatthew Berman · Forward FutureMatthew Berman tests Meta’s models on browser demos and reviews their published benchmark scores, availability and cost.모델과 학습BenchmarksAI cost & performance+1