- The Startup Club
- Posts
- š¶ļø AI safety or politics? Anthropicās quiet cleanup
š¶ļø AI safety or politics? Anthropicās quiet cleanup
Ex-OpenAI researcher subpoenaed in a copyright lawsuit
As AI reshapes industries, governments struggle to keep up. Anthropic erases safety pledges, Google expands AI search, and AI chatbots tweak their personalities. Meanwhile, the IRS cuts 45,000 jobs, and Justice cracks down on cyber threats.
The question isnāt just how AI evolves, but who gets to decide its futureācorporations, policymakers, or the algorithms themselves?
š„ Check out our NEW video: AI Has a Fatal FlawāAnd Nobody Can Fix It
ā¦

Seven bullets of updates
š IRS plans to cut up to 45,000 jobs, risking longer tax refund waits amid Trumpās federal downsizing push.
šµļø Justice charges 12 Chinese hackers for breaching 100+ US entities, including the Treasury; $10M reward offered.
š¬ Googleās AI "co-scientist" faces criticism for lacking real scientific utilityāreproducibility remains a key issue.
āļø Ex-OpenAI researcher subpoenaed in a copyright lawsuit; authors claim AI quoted their work without attribution.
š ļø Open Source Initiativeās board election sparks debate over the definition of "open source AI" amidst candidacy controversy.
š£ļø AI chatbots are becoming more extroverted, tweaking their personalities to seem more agreeableājumping from 50% to 95% extroversion.
š Google launches AI Mode in Search, enabling detailed, multi-part questionsāquery length doubles!
ā¦

Bret Taylor says AI agents will ruleāBelieve it?
At Mobile World Congress, OpenAI chair Bret Taylor made a strong case for AI agents, predicting they will become as significant as websites or mobile apps for brands. While avoiding a precise definition of "agentic AI," he highlighted their growing capabilities, particularly in customer service, where AI agents can provide multilingual, real-time assistance, reducing costs and enhancing user experiences. However, he acknowledged challenges, such as AI hallucinating policies that donāt exist, emphasizing the need for brand-specific safeguards. Taylor believes companies should focus on narrow applications of AI rather than waiting for the technology to be perfect.
Regarding the broader impact of AI, Taylor addressed job displacement, stating that while some roles will disappear, new opportunities will emerge. He called for public-private partnerships to ensure workforce reskilling keeps pace with AI-driven disruption. On OpenAIās shift from a nonprofit to a for-profit structure, Taylor insisted that its mission remains unchanged, but the cost of AI development necessitates funding adjustments. Looking ahead, he envisions AI agents fundamentally reshaping digital experiences, potentially reducing reliance on screens in favor of more seamless, conversational interactions.
ā¦

AI Has a Fatal FlawāAnd Nobody Can Fix It
AI is evolving fast, but what if there's a hard limit to how smart it can get? This video explores a fundamental equation that may cap AIās growth, revealing why even trillion-parameter models could be hitting a plateau.
From sky-high GPU costs to AIās struggles with reasoning and creativity, we break down the hidden roadblocks in its evolution š¶ļø Subscribe for more videos š¶ļø
ā¦

Start learning AI in 2025
Everyone talks about AI, but no one has the time to learn it. So, we found the easiest way to learn AI in as little time as possible: The Rundown AI.
It's a free AI newsletter that keeps you up-to-date on the latest AI news, and teaches you how to apply it in just 5 minutes a day.
Plus, complete the quiz after signing up and theyāll recommend the best AI tools, guides, and courses ā tailored to your needs.
ā¦

These companies just raised money
š³ Cino raises funding for its fintech app that lets friends and family split bills instantly and effortlessly.
š ļø Darwinbox secures $140M to take on Deel and Rippling with its SaaS platform for hiring, onboarding, and employee administration.
š° Grain emerges from stealth with $50M to help businesses hedge against FX volatility with finance-focused hedging tools.
š„¤ Olipop is now valued at $1.8B, solidifying its place as a leader in the better-for-you soda market.
š¦ Flex, a āBrex for business owners,ā raises $25M at a $250M valuation for its personal finance and payments software.
š· AMASS Brands is transforming the $900B beverage industry with a premium portfolio of natural, innovative drinks, from gin and tequila to non-alcoholic and organic wines.
ā¦

Hire an AI BDR & Get Qualified Meetings On Autopilot
Outbound requires hours of manual work.
Hire Ava who automates your entire outbound demand generation process, including:
Intent-Driven Lead Discovery Across Dozens of Sources
High Quality Emails with Human-Level Personalization
Follow-Up Management
Email Deliverability Management
ā¦

Anthropic hits delete on Bidenās AI safety pledges
Anthropic has quietly removed several Biden-era AI safety commitments, including pledges to share risk management research and address AI bias. These voluntary commitments, made in 2023 alongside OpenAI, Google, and others, were part of Bidenās push for ātrustworthyā AI but were never legally binding. The move follows Trumpās repeal of the AI Executive Order, which mandated AI safety testing and bias mitigation, citing concerns over excessive regulation.
Anthropic isnāt aloneāOpenAI has also scaled back DEI commitments, emphasizing intellectual freedom amid criticism of AI censorship. Trumpās new AI policy focuses on unrestricted development but drops anti-discrimination measures, signaling a broader shift in government oversight and industry compliance.
ā¦

Startup Events and Deadlines
Alchemist Accelerator | Deadline: March 21 | USA
Techstars Columbus | Deadline: April 1 | USA
Startup Grind Conference 2025 | April 19-30 | USA
Entrepreneurs Roundtable Accelerator | Deadline: April 28 | USA
Startup Battlefield 200 | Deadline: June 9 | Global
Entrepreneur First London - Summer | Deadline: July 1 | UK