- AI Emergence
- Posts
- ChatGPT Tasks: AI's Next Step into Personal Productivity
ChatGPT Tasks: AI's Next Step into Personal Productivity
Along with: AI’s Global Momentum: Bold New Plans from the U.S. and UK
Analytics Vidhya (Curated by Kunal Jain)
January 17, 2025
Hey there,
The race to lead in AI is on, and countries are moving fast to make their mark. The U.S. and the UK have rolled out big plans to build the energy, data centers, and tools needed to push AI innovation forward and stay ahead of the game.
While that’s happening on a global scale, OpenAI has taken a step toward bringing AI agents into our daily lives with their new Tasks feature. Have you tried it yet? What do you think? Drop a comment and let us know!
Now, let’s jump into this week’s updates.
What would be the format? Every week, we will break the newsletter into the following sections:
The Input - All about recent developments in AI
The Tools - Interesting finds and launches
The Algorithm - Resources for learning
The Output - Our reflection
Table of Contents
OpenAI is taking steady steps to introduce AI Agents into our daily lives. OpenAI has rolled out a new feature for ChatGPT called Tasks. This update allows users to schedule reminders and manage future actions, expanding its functionality beyond conversational interactions.
Key Features of the Tasks Update
Scheduling Made Simple
Users can now set one-time reminders or recurring tasks. Examples include receiving daily weather updates, weekly news briefings, or reminders for deadlines.Natural Language Prompts
Tasks leverage advanced natural language processing, making it easy to schedule actions using conversational commands like, “Remind me to send a report every Friday.”Personalized Suggestions
Based on previous interactions, ChatGPT can suggest relevant tasks, though user approval is required before they’re set.Integrated Management
Tasks can be managed through a dedicated section in the user profile or directly within chat threads, with notifications provided upon task completion.
Looking ahead, this development raises fascinating questions about the future of personal productivity tools. Could this mark the beginning of a new era where AI assistants become more deeply integrated into our daily routines? How might this shift our relationship with technology and task management? (source)
In a revealing chapter on AI's growing pains, Apple's ambitious foray into AI-generated news has hit a significant pause button. The tech giant's recent experience with its Apple Intelligence platform offers a fascinating glimpse into the delicate balance between innovation and reliability in AI deployment.
What makes this particularly noteworthy is the nature of the missteps - fabricated headlines about prime ministerial arrests and non-existent events that sent ripples through the journalism community. This situation illuminates a crucial challenge in AI development: the gap between algorithmic plausibility and factual accuracy.
Apple's response - temporarily disabling the feature and planning to add explicit AI content warnings - reflects a broader industry awakening to the complexities of automated news generation. The involvement of prestigious organizations like the BBC and advocacy groups adds weight to the ongoing dialogue about AI's role in information dissemination. (source)
President Biden has signed a key executive order to strengthen the U.S.’s leadership in AI. The plan focuses on building top-tier AI infrastructure and clean energy facilities, keeping America at the forefront of the global AI race while tackling security and environmental challenges.
AI Infrastructure: Building the Future
New AI Data Centers: Federal agencies are identifying prime locations for AI-ready data centers and clean energy sites.
Private-Sector Collaboration: Companies will bid to develop these facilities, with expedited permits and grid connections to fast-track progress.
Sustainability Focus: Developers must adhere to environmental standards, ensuring data centers are powered sustainably without raising energy costs.
This move signals that U.S. is serious about being self-reliant and not to outsource key AI related opportunities outside.
Chips and Allies: A Tiered Export Policy
Alongside the executive order, Biden introduced new restrictions on exporting advanced AI chips to safeguard national security. Countries are now categorized into three tiers:
Tier 1 (Allies): Nations like Japan, Canada, and the UK enjoy full access to advanced AI chips.
Tier 2 (Limited Access): Countries such as Israel, Mexico, and Switzerland face purchase limits.
Tier 3 (Adversaries): China, Russia, and others remain blocked from accessing these chips.
This tiered approach is designed to prevent sensitive technology from falling into the wrong hands while strengthening partnerships with trusted allies.
The executive order and export restrictions have sparked mixed reactions across the tech industry:
Concerns from Nvidia: The company called the restrictions “misguided,” warning they could stifle innovation and disrupt global supply chains.
Pushback from SIA: The Semiconductor Industry Association expressed disappointment, citing fears of long-term damage to U.S. competitiveness in strategic markets.
On January 12, 2025, the UK introduced its AI Opportunities Action Plan, an initiative to boost economic growth in the AI boom. Backed by a £14 billion investment from top tech companies and the potential to create over 13,000 jobs, the plan aims to cement the UK’s position as a global leader in AI innovation.
Key Highlights of the AI Opportunities Action Plan
AI Across Sectors
The plan proposes to bring AI into different sectors, making them more efficient and productive. AI applications will be used to:Improve healthcare services.
Streamline public transport systems.
Assist local governments with tasks like identifying road potholes.
AI Growth Zones (AIGZs)
A major feature of the initiative is the establishment of AI Growth Zones, designed to fast-track AI infrastructure development. These zones will:Provide enhanced power access.
Expedite planning approvals.
Launch with the first AIGZ in Culham, Oxfordshire, home to the UK Atomic Energy Authority. This site will serve as a testing ground for public-private partnerships, including one of the UK’s largest AI data centers.
Supercharged Computing Power
To support cutting-edge AI research, the UK plans to increase public computing capacity by 20x, starting with the development of a new supercomputer designed for advanced AI processing.National Data Library
The plan includes creating a National Data Library to safely unlock public data for AI development while ensuring data security and privacy.AI Energy Council
A dedicated AI Energy Council will collaborate with energy companies to manage the power demands of AI technologies. This aligns with the UK’s vision of becoming a clean energy superpower. (source)
In an exciting leap for agentic AI, AutoGen v0.4 redefines scalability, extensibility, and robustness in multi-agent applications. Building on community feedback, this release introduces a redesigned architecture with standout features like asynchronous messaging, modular customization, and advanced observability tools.
Users can now design distributed agent networks with cross-language support, leveraging Python and .NET, and extend functionality through community-driven integrations.
The update enhances developer tools with AutoGen Bench for benchmarking and AutoGen Studio for low-code prototyping. Studio offers real-time updates, mid-execution control, and intuitive visualization for managing agent workflows. New capabilities include drag-and-drop team building and Magentic-One, a multi-agent system tackling complex tasks across domains. (source)
In a fascinating development that highlights the complex intersection of AI innovation and intellectual property rights, Meta finds itself at the center of a thought-provoking legal controversy. The case brings to light crucial questions about how tech giants build their AI models and the ethical boundaries of data usage in the age of artificial intelligence.
What makes this particularly intriguing is the personal dimension: renowned authors Sarah Silverman and Ta-Nehisi Coates stepping forward to challenge one of tech's most powerful companies. Their allegations that Meta, under Mark Zuckerberg's direction, deliberately used pirated datasets like LibGen for AI training adds a compelling layer to the ongoing dialogue about AI ethics and accountability.
The controversy touches on a deeper narrative about the balance between technological progress and creative rights. Meta's fair use defense raises fascinating questions about how we define and protect intellectual property in an era where AI training requires vast amounts of data. (source)
In a fascinating development that could reshape the AI landscape, UC Berkeley's Sky Computing Lab has achieved what many thought impossible. Their NovaSky research team has unveiled Sky-T1-32B-Preview, a breakthrough that challenges our assumptions about the cost barriers in advanced AI development.
What makes this particularly intriguing is the economics: a mere $450 to train a sophisticated 32-billion-parameter model. To put this in perspective, similar models typically required investments in the millions - a cost reduction that could democratize AI research in unprecedented ways.
The model's development story showcases the power of collaborative innovation. By leveraging synthetic data curated with help from existing models like Alibaba's QwQ-32B-Preview and OpenAI's GPT-4o-mini, the team demonstrated how smart resource utilization can yield remarkable results. Training took just 19 hours on 8 Nvidia H100 GPUs - a testament to efficient engineering and creative problem-solving.
What's particularly noteworthy is Sky-T1's performance profile. While it excels in mathematics and coding challenges, outperforming some established benchmarks, its relative weakness in graduate-level science problems reveals the nuanced nature of AI capabilities. This "fact-checking" ability in its problem-solving approach suggests an interesting evolution in AI reasoning capabilities. (source)
X has taken a bold step by launching Grok AI as a standalone app on Apple's App Store. This move reflects an intriguing evolution in how social platforms are integrating AI capabilities into their ecosystem.
X's strategic approach to AI deployment is noteworthy here. By offering a free tier with 10 requests per hour alongside an $8 monthly Premium plan, they're experimenting with a hybrid model that balances accessibility with sustainability.
The app's focus on "humorous and engaging" interactions, combined with practical features like image generation and text summarization, suggests a deliberate attempt to differentiate from more conventional AI assistants. (source)
OpenAI has unveiled its Economic Blueprint, outlining policies to ensure AI drives equitable economic growth, strengthens national security, and maintains U.S. global leadership in innovation. Key highlights include:
Maximizing AI's Benefits: Proposals focus on creating jobs, expanding AI infrastructure (e.g., data centers, chip plants), and ensuring shared prosperity across communities.
Learning from History: Drawing parallels to the automobile revolution, OpenAI emphasizes the need for unified national standards to unlock AI’s full potential while safeguarding public safety.
Clear Rules for AI: Common-sense, nationwide regulations are proposed to encourage investment, foster competition, and protect democratic values.
Key Focus Areas:
Strengthening resources like chips, data, energy, and talent.
Promoting free-market innovation while safeguarding citizens from government misuse.
Attracting global AI investments to maintain U.S. leadership over competitors like China. (source)
Google is fundamentally transforming the digital workspace by seamlessly weaving AI capabilities throughout its Workspace platform. The reimagined Business and Enterprise plans now offer integrated AI tools - from Gemini's sophisticated assistance in Gmail and Docs to NotebookLM Plus's collaborative learning features - all without complex add-ons.
This comprehensive approach, backed by robust security and competitive pricing, signals a thoughtful evolution in how we work, making advanced AI capabilities accessible while prioritizing practical productivity gains. The result is a unified ecosystem where AI naturally enhances existing workflows, suggesting a deeper understanding of how technology can meaningfully support modern business needs. (source)
Adobe has unveiled new generative AI tools, including Firefly Bulk Create, which allows users to edit up to 10,000 images in a single click. Designed to simplify production tasks, this tool enables background removal, resizing for platforms like TikTok and Instagram, and bulk image customization. Firefly Bulk Create, now in beta, integrates Adobe’s APIs for user-friendly creative workflows. Additional AI tools include video dubbing in 14 languages, auto-formatting for media, and digital avatars, with usage tied to a premium plan offering generative credits. (source)
ChatGPT’s Scheduled Tasks Feature
ChatGPT now offers a ‘Scheduled Tasks’ feature that lets users automate prompts and receive timely responses or notifications. Imagine having ChatGPT prepare your daily newsletter, organize your work schedule, or remind you of important events like birthdays or meetings—all at pre-set times. This feature works seamlessly to send you updates via push notifications or email, even when you're offline!
How to Access
1. Access the Scheduled Tasks Feature
Open ChatGPT.
Click your profile icon in the top-right corner.
Select “Tasks” from the dropdown menu.
2. Create a New Task
Select “Try your own” or “ChatGPT with scheduled tasks” from the model selector.
In the chat window, type your prompt and schedule. Examples:
“Send me a summary of top news articles every morning at 7.”
Once you submit, you’ll see a confirmation message saying “Task saved.”
3. Edit, Pause, or Delete Tasks
Navigate to the three-dot menu next to the task.
Choose “Edit”, “Pause”, or “Delete” as needed.
4. Enable Notifications
Go to “Notification Settings” in ChatGPT Web.
Turn on push notifications, email notifications, or both.
For desktop, allow notifications via browser settings.
On mobile, grant permissions when prompted after creating a task.
5. Organize Your Tasks
Tasks are run within the same chat window where they are set up.
To create separate tasks with unique notifications, start a new conversation for each.
Dive into the world of GenAI with this free course on building LLM applications- Building LLM Applications using Prompt Engineering. You'll master prompt engineering, implement best practices, and develop chatbots using enterprise data. Explore advanced techniques like few-shot and one-shot prompting to enhance your skills and create impactful applications.
Also, check out this latest episode of The Joe Rogan Experience episode with Mark Zuckerberg, where he explores AI, neural interfaces, and AR. He discusses early medical uses of neural tech, like restoring sensation, and future everyday applications. Zuckerberg envisions AR glasses enhancing communication, AI automating routine tasks, and open-source AI driving innovation, empowering humanity, and fostering connection.
The year’s just begun, and the AI race is already picking up speed!
What do you think about the latest AI plans from the US and UK? Who’s going to come out on top as the AI superpower in the next five years? Let us know in the comments.
Exciting times ahead!
Reply