Are AI Platforms Like ChatGPT Overpriced? My Scoring System Says Yes

As a Red investor, I’m skeptical of the hype driving sky-high valuations for AI platforms like OpenAI’s ChatGPT, Anthropic’s Claude, xAI’s Grok, Microsoft’s Copilot, and Alphabet’s Gemini. With ChatGPT boasting 700 million weekly users and OpenAI valued at $500 billion, the AI boom is real—but are these platforms worth their price tags?

I’ve built a scoring system (1–100) to cut through the noise, using company metrics to determine if these platforms are overpriced. Spoiler: Most are, with valuations outpacing fundamentals.

My AI Platform Scoring System: How It Works

I’ve designed a 1–100 scoring system to evaluate AI platforms based on five key categories, weighted to balance growth potential with bubble risks. It’s a no-nonsense way to see if valuations match reality across the market. Here’s the breakdown:

• Valuation Metrics (30%): Compares forward P/E, P/S, or implied revenue multiples to the software industry average (5.7x P/S). Crazy multiples like 25x+ signal overpricing.
• Revenue Growth and Profitability (25%): Measures ARR/revenue growth and profit outlook. High growth is great, but billion-dollar losses (e.g., OpenAI’s $14B projected 2025 loss) are a red flag.
• Market Position and User Adoption (20%): Looks at user base and market share. Leaders like ChatGPT score high; niche players like Grok lag.
• External Risks (15%): Factors in tariffs (e.g., chip restrictions costing $5.5B), regulatory scrutiny (e.g., antitrust), and economic pressures. Higher risks, lower scores.
• Sentiment and Momentum (10%): Tracks analyst ratings and X sentiment. Hype inflates scores; sell-off fears drag them down.

Each category gets a 0–100 score, weighted to produce a final score: 80–100 (strong buy), 60–79 (stable), 40–59 (caution), 20–39 (weak), 1–19 (recession risk). I’ve scored five platforms using September 2025 data from Reuters, CNBC, and Morningstar, considering market-wide dynamics.

Scoring the AI Platforms1. OpenAI (ChatGPT)

• Valuation Metrics (30%): $500B valuation, $13B ARR (Q3), projected $20B by December. Implied P/S: 25–38x (vs. 5.7x software avg). No profits, $14B loss projected for 2025. Score: 25/100 (way overpriced).
• Revenue Growth and Profitability (25%): 100% YoY ARR growth ($3.7B to $13B), but $5B 2024 loss, $14B 2025 loss due to compute costs. Score: 60/100 (stellar growth, brutal losses).
• Market Position and User Adoption (20%): 700M weekly users, 74.8% AI search share, 10M paying subscribers, 5M business users. Score: 90/100 (king of the hill).
• External Risks (15%): Heavy Azure reliance, tariff-driven chip costs, $40B funding needs. Regulatory scrutiny looms. Score: 40/100 (big risks).
• Sentiment and Momentum (10%): Mixed X sentiment (hype vs. bubble fears), $40B SoftBank-led raise. Score: 60/100 (optimism fading).
• Total Score: (25 × 0.3) + (60 × 0.25) + (90 × 0.2) + (40 × 0.15) + (60 × 0.1) = 52.5/100 (Caution). ChatGPT’s dominance is real, but $500B is absurd with massive losses.

2. Anthropic (Claude)

• Valuation Metrics (30%): $170B valuation, $3.5B ARR. P/S: ~48x. Unprofitable, no P/E. Score: 20/100 (insane multiple).
• Revenue Growth and Profitability (25%): ~100% YoY ARR growth, $5B+ losses. Score: 55/100 (strong growth, big losses).
• Market Position and User Adoption (20%): ~100M users, 10–20% AI search share. Growing via AWS. Score: 60/100 (solid but no ChatGPT).
• External Risks (15%): AWS dependency, tariff costs, $5B raise needs. Antitrust risks. Score: 35/100 (high risks).
• Sentiment and Momentum (10%): Funding buzz, but X calls it “second-tier.” Score: 55/100 (mixed vibes).
• Total Score: (20 × 0.3) + (55 × 0.25) + (60 × 0.2) + (35 × 0.15) + (55 × 0.1) = 42.5/100 (Caution). Claude’s growth can’t justify $170B with losses and risks.

3. xAI (Grok)

• Valuation Metrics (30%): $24B valuation, no ARR data. Speculative multiple, unprofitable. Score: 30/100 (less frothy but unclear).
• Revenue Growth and Profitability (25%): Early-stage, no ARR, unprofitable. High potential. Score: 40/100 (unproven).
• Market Position and User Adoption (20%): <10M users, <5% share. Niche truth-seeking focus. Score: 20/100 (small player).
• External Risks (15%): Compute costs, tariffs, funding reliance. Regulatory uncertainty. Score: 30/100 (severe risks).
• Sentiment and Momentum (10%): X loves “truth-seeking” AI, no analyst targets. Score: 50/100 (neutral).
• Total Score: (30 × 0.3) + (40 × 0.25) + (20 × 0.2) + (30 × 0.15) + (50 × 0.1) = 32.5/100 (Weak). Grok’s lower valuation has upside, but it’s too early to bet big.

4. Microsoft (Copilot)

• Valuation Metrics (30%): $3T market cap, $83.3B Azure revenue (proj. 2025). P/E: 35x (vs. S&P 500 22.9x). Azure P/S: ~36x. Score: 60/100 (premium but diversified).
• Revenue Growth and Profitability (25%): Azure +28.6% YoY, 34% net margin. Score: 85/100 (strong and profitable).
• Market Position and User Adoption (20%): Copilot in 1B+ Office users, Azure AI leader. Score: 90/100 (top-tier).
• External Risks (15%): Moderate tariff/antitrust risks, manageable economic impact. Score: 70/100 (manageable).
• Sentiment and Momentum (10%): Strong buy ratings, positive X buzz. Score: 80/100 (bullish).
• Total Score: (60 × 0.3) + (85 × 0.25) + (90 × 0.2) + (70 × 0.15) + (80 × 0.1) = 75.75/100 (Moderate). Copilot’s integration and profits make it fairly priced.

5. Alphabet (Gemini)

• Valuation Metrics (30%): $2.5T market cap, no Gemini revenue. P/E: 25x. P/S: ~6x. Score: 75/100 (reasonable).
• Revenue Growth and Profitability (25%): Google Cloud +15% YoY, 26% net margin. Score: 80/100 (steady, profitable).
• Market Position and User Adoption (20%): 450M Gemini users, 20–30% AI search share. Score: 80/100 (strong but trails ChatGPT).
• External Risks (15%): Antitrust, moderate tariffs, economic pressures. Score: 65/100 (moderate risks).
• Sentiment and Momentum (10%): Buy ratings, positive X sentiment. Score: 75/100 (solid).
• Total Score: (75 × 0.3) + (80 × 0.25) + (80 × 0.2) + (65 × 0.15) + (75 × 0.1) = 75.75/100 (Moderate). Gemini’s diversified revenue keeps it grounded.

My Take: Most AI Platforms Are Overpriced

My scoring system confirms what I suspected: AI platforms like ChatGPT (52.5, Caution), Claude (42.5, Caution), and Grok (32.5, Weak) are overpriced, with insane P/S ratios (25–48x vs. 5.7x software average) and massive losses ($5–14B). Microsoft’s Copilot and Alphabet’s Gemini, both scoring 75.75 (Moderate), are exceptions, with diversified revenue and profits making them fairly priced. X posts echo my concerns, calling OpenAI’s $500B valuation a “bubble” when training costs ($1B/model) dwarf subscriptions ($20/month). Tariffs (e.g., H20 chip restrictions costing $5.5B) and antitrust risks add pressure. I’m steering clear of private platforms like OpenAI and Anthropic until valuations crash, maybe post-IPO. For now, I’m eyeing broad-market AI ETFs like Global X AIQ to spread risk across tech, industrials, and defensives.