Last update: Monday 3/17/25
ChatGPT's TL;DR summaries of Top 2 stories ...
Introduction:
Our podcast "Introduction" will discuss why Siri cannot be "fixed" by an "upgrade". Replacing iOS 18.2 with iOS 18.3, was an upgrade that fixed important, but relatively small problems. By contrast, replacing any iOS with an iOS that would enable Siri to provide services that are comparable to the services provided by industry leading assistants like ChatGPT or Claude or Copilot or Gemini, would not be an "upgrade"; it would be a technology revolution because Siri was created way back in 2011 using "ancient" machine learning technology.
Siri cannot become comparable to other AI assistants until its underlying "ancient" AI technology is replaced by the world changing generative AI technology that OpenAI first announced to the general public in November 2022.
1) Intense reactions within the Apple communiity to Apple's delay of Siri upgrades
-- "Something Is Rotten in the State of Cupertino", John Gruber, Daring Fireball, 3/12/25 -- "Apple’s Siri Chief Calls AI Delays Ugly and Embarrassing, Promises Fixes", Mark Gurman, Bloomberg, 3/14/25
-- Subsequent reactions reported by MacRumors, 9t05Mac, The Verge, Gizmodo,
PROMPT:
Here comes the text for two articles. Provide concise summaries of each article, presenting your findings in bullets. Both summaries should be understandable by savvy computer users.
- Text Daring Fireball
- Twxt Bloomberg
RESPONSE:
1. Daring Fireball: “Something Is Rotten in the State of Cupertino”
- Apple’s credibility crisis:
The article argues that Apple has severely damaged its long-standing reputation for delivering what it promises by prematurely announcing and marketing AI-powered Siri enhancements that were far from ready. - Failure to demo real features:
At WWDC 2024, Apple showcased “personalized Siri” features using a concept video rather than live demonstrations. Unlike past Apple announcements, no actual working version of these features was ever shown. - Marketing blunder:
Apple doubled down on promoting these AI-powered Siri capabilities in iPhone 16 commercials despite knowing they were nowhere near completion. The commercial was later quietly removed after the public delay announcement. - Hierarchy of feature readiness:
The article categorizes software releases into four levels (from demoed live to shipped), pointing out that the missing Siri features were level 0—vaporware. Apple never allowed journalists to test them, which should have raised red flags. - WWDC deception:
Apple historically only showcased products that were at least close to shipping. The article compares this situation to Apple’s 1987 “Knowledge Navigator” concept video, which foreshadowed the company’s near-bankruptcy in the ’90s. - Executive accountability:
The decision to market unfinished AI features was a leadership failure, with Tim Cook and his top executives responsible for misleading announcements and poor planning. - Comparison to Steve Jobs era:
The author recalls how Steve Jobs publicly berated teams for failing to deliver (e.g., MobileMe in 2008), suggesting that Apple’s current leadership lacks similar accountability. - Biggest issue:
The real problem isn’t that Apple is behind in AI—it’s that Apple lied about its progress, damaging the trust it built over decades.
2. Bloomberg: “Apple’s Siri Chief Calls AI Delays Ugly and Embarrassing, Promises Fixes”
- Internal backlash: Robby Walker, Apple’s Siri senior director, admitted in an internal meeting that the delays were “ugly and embarrassing.”
- Marketing before readiness:
Apple publicly promoted features that weren’t functional, making the failure worse when they had to backtrack. - Engineering struggles:
Apple has faced serious technical challenges, with AI enhancements originally planned for spring 2025 now postponed to iOS 19 (2026). - Failed expectations:
The delayed features include personal context awareness, on-screen understanding, and advanced in-app actions—all essential to making Siri competitive with modern AI assistants. - Executives under pressure:
Craig Federighi (software), John Giannandrea (AI), and other leaders are now facing scrutiny over the botched rollout. - Stock impact:
Apple’s AI delays have contributed to a 16% drop in its stock price this year. - Doubts on future timeline:
Even the revised iOS 19 target is uncertain, as Apple must prioritize other projects, potentially delaying AI advancements even further. - Damage control:
Walker promised that Apple is still working on making Siri a world-class AI assistant, but did not confirm a concrete release timeline.
2) Responses to NSF/OSTP request for input on AI Action Plan
-- "OpenAI’s proposals for the U.S. AI Action Plan", OpenAI, 3/13/25-- "Response to the National Science Foundation’s and Office of Science & Technology Policy’s Request for Information", Google, 3/13/2
-- Response to "Request for Information (RFI) on the Development of an Artificial Intelligence (AI) Action Plan (“Plan”)", Anthropic, 3/8/25
-- "Response to the National Science Foundation’s and Office of Science & Technology Policy’s Request for Information on the Development of an Artificial Intelligence (AI) Action Plan T", News Media Alliance, 3/14/25
PROMPT:
Here come text for 4 policy proposals to the National Science Fooundation and the Office of Science and Technology Policy. The first 3 proposals come from GenAI providers -- Anthropic, Google, and OpenAI. The fourth proposal comes from the News Media Alliance.
Provide a detailed summary of each of these 4 proposals in a no nonsense style, with key points presented in bullets for easier reading;
Then write a concise summary of the points that the three proposals from Anthropic, Google, and OpenAI are in agreement with each other, with key points presented in bullets. Label this summary the "GenAI Providers Consensus".
Finally compare your "GenAI Providers Consensus" with your summary of the proposal from the News Media Alliance. Identify the points on which they agree and the points on which they disagree with each other. Present your comparisons in bullets.
All summaries should be understandable by savvy computer users.
This is a complex task. If you have any questions about what is requested, ask me those questions before you proceed.
- Text Anthropic
- Text Google
- Text OpenAI
- Text News Media Alliance
RESPONSE:
Summaries of the Four Policy Proposals
1. Anthropic’s AI Action Plan
- National Security & AI Leadership
-- AI should be treated as a critical national asset with government-backed safeguards.
-- AI development should bolster American economic competitiveness and security.
-- Future AI systems could match or surpass Nobel-level intelligence across multiple fields.
-- AI models must be able to interact with digital and physical environments autonomously. - Government Oversight & AI Safety
-- The U.S. should test and evaluate AI models for potential national security threats.
-- The Department of Commerce’s AI Safety Institute should expand AI model testing partnerships.
-- NIST should develop security protocols to detect AI-related threats before they materialize.
- Export Controls & Cybersecurity
-- Strengthen export controls on chips and AI model weights to prevent adversarial use.
-- Limit access to AI infrastructure to aligned nations through government-to-government agreements.
-- Increase funding for cybersecurity and establish classified communication channels between AI labs and intelligence agencies.
- Energy & Infrastructure Expansion
-- AI will require a massive increase in U.S. energy supply (50 GW by 2027).
-- Accelerate government-backed infrastructure projects to meet AI energy needs.
-- Expedite permitting for data centers, power plants, and grid expansions.
- Government AI Adoption
-- Implement AI in federal agencies to automate workflows and increase efficiency.
-- Use AI to streamline operations in the IRS, VA, and Social Security Administration.
-- Modernize procurement processes to acquire AI-driven systems more effectively.
2. Google’s AI Policy Proposal
- Investment in AI
-- Increase federal investment in AI R&D and provide more compute resources for researchers.
-- Support energy transmission reform to scale data centers and AI infrastructure.
-- Balance export controls to protect U.S. security without harming AI exports.
- AI Regulation & Federal Governance
-- Federal AI regulation should be uniform to avoid a patchwork of conflicting state laws.
-- AI should operate under risk-based governance models rather than blanket regulations.
-- Copyright, privacy, and patent laws should ensure AI access to training data.
- Government AI Modernization
-- The federal government should use AI in public administration, streamlining services.
-- AI procurement rules should favor multi-cloud, multi-vendor approaches.
-- Agencies should adopt automated monitoring & AI security protocols.
- International AI Policy
-- Promote American-led AI standards in global markets.
-- Challenge foreign AI barriers that limit U.S. exports or promote protectionist policies.
-- Develop reciprocal AI security agreements with allied nations.
3. OpenAI’s AI Policy Proposal
- AI as a Strategic Asset
-- The U.S. must maintain dominance in AI to counter China’s AI ambitions.
-- AI should enhance freedom, prosperity, and security by scaling human ingenuity.
- Government-AI Partnership
-- Voluntary AI partnerships between companies and government should replace state-level AI regulations.
-- The U.S. Commerce Department should oversee national AI security coordination.
- Export Controls & AI Leadership
-- Use export controls to prevent AI diffusion to China and adversaries.
-- Promote U.S.-aligned AI in global markets to counter China’s state-controlled AI.
-- Expand infrastructure & transmission projects to ensure AI-powered reindustrialization.
- Copyright & Fair Use
-- AI companies should have broad access to training data under fair use.
-- U.S. should oppose restrictive copyright policies like the EU’s opt-out model.
-- AI-generated content should be clearly labeled but not hinder model training.
- Infrastructure & National Security
- -- Public-private investment should accelerate AI data center expansion.
-- AI-ready workforce training must be a national priority.
-- Defense-specific AI models should be developed for classified applications.
4. News Media Alliance (N/MA) Proposal
- Intellectual Property & AI Licensing
-- AI models must compensate publishers for using copyrighted content.
-- Fair market licensing is necessary to ensure publishers benefit from AI monetization.
-- Opposes the AI industry’s use of copyrighted material without permission.
- Transparency & Fair Competition
-- AI companies must disclose how content is used in AI models.
-- Supports tracking mechanisms to identify AI-scraped content.
-- More competition protections to prevent Big Tech from monopolizing AI-powered search.
- Opposition to Free Use of Copyrighted Content
-- Criticizes the AI industry’s stance on fair use, arguing it undermines content creators.
-- Rejects OpenAI’s copyright position, calling for clear licensing agreements.
-- Opposes foreign “opt-out” AI data mining laws that disadvantage U.S. publishers.
GenAI Providers Consensus (Anthropic, Google, OpenAI)
- National AI Strategy
-- The U.S. must maintain AI dominance for economic and national security reasons.
-- AI models should be treated as critical assets, requiring government oversight.
- Export Controls & National Security
-- Support export controls on AI chips and models to limit China’s AI capabilities.
-- The U.S. must lead in global AI standardization to align AI governance with democratic values.
- Government Investment & AI Adoption
-- Expand AI infrastructure, including energy and data centers.
-- Encourage AI adoption in government services to increase efficiency.
- Balanced AI Regulation
-- AI regulation should be federal, not state-based, to prevent fragmented laws.
-- Supports risk-based governance models instead of blanket AI restrictions.
- Intellectual Property & Copyright
-- AI training should retain access to copyrighted material under fair use.
-- Opposes restrictive copyright laws that hinder AI model development.
Comparison: GenAI Providers vs. News Media Alliance
Agreements
- AI is vital for U.S. economic and national security leadership.
- Transparency in AI development is necessary, though motivations differ.
- AI infrastructure & investment should be prioritized to maintain competitiveness.
Disagreements
- Intellectual Property & Copyright
-- GenAI Providers: AI should use copyrighted content under fair use to ensure model quality.
-- N/MA: AI companies must license copyrighted materials and compensate publishers. - Market Regulation
-- GenAI Providers: Favor free-market AI development with minimal government intervention.
-- N/MA: Supports stricter AI content licensing and government oversight to protect publishers. - Transparency
-- GenAI Providers: Prefer voluntary transparency but oppose mandatory disclosure of AI training data.
-- N/MA: Supports mandatory AI disclosure policies to track copyright infringement.
- AI’s Impact on Competition
-- GenAI Providers: Advocate for open AI markets to maximize competition.
-- N/MA: Concerned that Big Tech’s AI dominance could stifle smaller publishers.
Conclusion
- GenAI providers prioritize AI expansion, national security, and minimal regulation.
- News Media Alliance focuses on protecting intellectual property, fair compensation, and competition fairness.
No comments:
Post a Comment
Your comments will be greatly appreciated ... Or just click the "Like" button above the comments section if you enjoyed this blog note.