
Bad AI: Zucker Punched
Apr 17, 2025
AI IS ACCELERATING - GET READY
Let's help you understand the WTF of AI and how it helps YOU and your organisation grow. Ready, set, agent.
This week:
π± Bad AI: Zucker Punched
π AI Moves: Claude Research
π₯ AI Masterclass: For Senior Marketers
β¨ AI Remembers: ChaGPT Memory Update
π‘ Trends & Insights
BAD AI: Zucker Punched
![]() |
This week in “AI is definitely making society better,” someone thought: You know what the world needs? Elon Musk and Mark Zuckerberg spouting tech nonsense at us while we cross the street.
β
Yep. Silicon Valley crosswalk buttons were changed to play AI-generated imitations of our human/tech overlords. Pedestrians were serenaded with phrases like “Cross now or face the singularity” in Musk’s child-voice, and “Put these Metaverse glasses on before you walk” in Zuckerberg’s emotionally unavailable, lizard-man monotone.
β
Back in the valley, tech-bro locals reportedly preferred these tech voices to the standard crossing voice. One startup founder described it as “Disruptive pedestrian codex for UX".*
β
Of course they did, the VC-entitled, tri-latte drinking tw*t. (*still being fact-checked as we went to print).
AI MOVES: Claude Research
![]() |
Anthropic has launched a suite of early beta features that push Claude (its LLM) closer to being a functional AI research assistant. The update includes a new multi-step research mode, deeper integration with Google Workspace, and enterprise-level document indexing.
β
Key Highlights:
β
- Research Mode: Claude can now autonomously break down a query into sub-questions, run multiple live searches, and return structured responses with citations. Early use cases: competitive analysis, trend scanning, market mapping.
- Google Workspace Integration: Claude connects to Gmail, Docs, and Calendar with user consent - enabling it to extract context, summarise threads, prep for meetings, or surface relevant files without uploading or re-prompting.
- Enterprise Knowledge Indexing: Companies can index internal documents (e.g. wikis, decks, PDFs) to retrieve - using secure retrieval-augmented generation (RAG) techniques.
- No file uploads needed: Retains session context and workspace awareness, removing a bit of friction in daily use.
β
What it means for teams:
Claude is getting better equipped for real work - reducing time spent gathering intel, navigating internal docs, or re-summarising the same threads.
β
The document indexing and email access can help operate inside your workflow,s not just respond to questions in a vacuum. Still early beta, but the direction is practical and highly relevant for comms, research and marketing teams.
β
AI workflows are changing rapidly. Talk to us about the most effective AI research tools and processes for your organisation.
AI MASTERCLASS: For Senior Marketers
|
Join a select group of top marketers and unlock AI’s power for research, competitor insights, content, and personalisation at scale.
β
Senior Marketers AI Masterclass:β
Thursday 15th May, 10.00am – 11.00am BST.
β
What you get:
β
β
The highest impact areas to use AI within marketing.
β The latest, best AI tools to use every day.
β Practical, relevant use cases - knowledge you can wow your team with.
β A plan on how to transform your marketing function over the coming year.
β
Ready to unlock real competitive advantage with AI? Apply now to reserve your place – limited spots available.
β
π Don't wait, reserve your place here.
AI REMEMBERS: ChatGPT Memory
|
OpenAI has switched on automatic memory for ChatGPT - making it the first major assistant with persistent, cross-session recall by default. It remembers who you are, what you’ve said, and how you like to be answered - with no prompt engineering required.
β
Key Takeaways:
β
- Automatic Memory Activation: ChatGPT now stores information from your interactions by default, personalising future responses without the need for manual prompts or reminders.
- Comprehensive Recall: The assistant remembers facts you share, your communication style, and previous topics discussed - unless you explicitly turn it off.
- User Control: Users can view, edit, or delete individual memories, or disable the memory feature entirely through the settings menu.
- Privacy Considerations: Unless opted out, user data may be utilised to train OpenAI's broader models, not solely for personal assistant improvements.
β
What it means for Businesses:
This approach shifts AI from one-off tool to long-term assistant. Businesses need clear policies, training on memory management, and controls on what data is being stored and shared.
β
Context with other LLM memory:
Claude doesn't retain information between sessions (safer, potentially less helpful). Gemini pulls context quietly from your Google tools. ChatGPT is the most persistent - and the most aggressive. It's powerful, but needs oversight. Also, Grok aren't waiting around either.
β
If your organisation needs clarity on how AI fits into your business, we help teams adopt AI tools with the right balance of capability, control, and compliance. Get in touch.
WHATS NEW - Insights & Trends
β
> Is US AI innovation suffering?
> Open AI safety adjustmentsβ
> Why is everyone a doll?
> Microsoft Total Recall is backβ
> Personified AI is helping peopleβ
G3NR8 is an impact consultancy for the exponential age. We help organisations figure out the highest impact areas to use AI for maximum impact.
β
Work with us to:
- Develop your AI strategy and policies.
- Create a competitive advantage using your proprietary data.
- Automate repetitive tasks.
- Gain actionable insight from data.
- Train your teams - we've trained 150 large organisations in advanced generative principles and implementation.
β
βDrop us a line for a chat about productivity gains, cost savings and how AI can power growth in your business.
β
Stay Ahead with The Ultimate AI Newsletter
Subscribe forΒ unique AI insightsΒ andΒ strategiesΒ that redefine business and innovation. Plus, get VIP access to a curated selection of "bad AI" - because sometimes, learning what not to do is just as valuable..